RTI Routing Service

User’s Manual

Version 5.0

© 2009-2012 Real-Time Innovations, Inc.

All rights reserved.

Printed in U.S.A. First printing.

August 2012.

Trademarks

Real-Time Innovations, RTI, and Connext are trademarks or registered trademarks of Real-Time Innovations, Inc. All other trademarks used in this document are the property of their respective owners.

Copy and Use Restrictions

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form (including electronic, mechanical, photocopy, and facsimile) without the prior written permission of Real- Time Innovations, Inc. The software described in this document is furnished under and subject to the RTI software license agreement. The software may be used or copied only under the terms of the license agreement.

Technical Support

Real-Time Innovations, Inc.

232 E. Java Drive

Sunnyvale, CA 94089

Phone:

(408) 990-7444

Email:

support@rti.com

Website:

https://support.rti.com/

Contents

1

Welcome to RTI Routing Service

 

 

1.1

Available Documentation .........................................................................................................................

1-3

2

Configuring Routing Service

 

 

2.1

Terms to Know ...........................................................................................................................................

2-1

 

2.2

How to Load the XML Configuration ....................................................................................................

2-2

 

2.3

XML Syntax and Validation .....................................................................................................................

2-3

 

2.4

XML Tags for Configuring Routing Service ..........................................................................................

2-5

 

 

2.4.1

Routing Service............................................................................................................................

2-6

 

 

2.4.2

Domain Route ..............................................................................................................................

2-9

 

 

2.4.3

Administration...........................................................................................................................

2-13

 

 

2.4.4

Monitoring..................................................................................................................................

2-15

 

 

2.4.5

Session.........................................................................................................................................

2-20

 

 

2.4.6

Routes..........................................................................................................................................

2-22

 

 

2.4.7

Auto Routes................................................................................................................................

2-32

 

 

2.4.8

Adapters .....................................................................................................................................

2-37

 

2.5

Enabling and Disabling Routing Service Entities ...............................................................................

2-37

 

2.6

Enabling RTI Distributed Logger in Routing Service.........................................................................

2-38

 

2.7

Support for Extensible Types .................................................................................................................

2-39

 

 

2.7.1

Example ......................................................................................................................................

2-40

3

Running Routing Service

 

 

3.1

Starting Routing Service ...........................................................................................................................

3-1

 

3.2

Stopping Routing Service .........................................................................................................................

3-1

4

Transforming Data with Routing Service

 

 

4.1

Transformation Usage and Configuration .............................................................................................

4-1

 

4.2

Transformations Distributed with Routing Service..............................................................................

4-3

 

4.3

Creating New Transformations ...............................................................................................................

4-4

 

 

4.3.1

Transformation Plugin API........................................................................................................

4-5

5

Administering Routing Service from a Remote Location

 

 

5.1

Enabling Remote Administration ...........................................................................................................

5-1

iii

 

5.2 Remote Commands ...................................................................................................................................

5-2

 

 

5.2.1

add_peer .......................................................................................................................................

5-3

 

 

5.2.2

create .............................................................................................................................................

5-3

 

 

5.2.3

delete .............................................................................................................................................

5-4

 

 

5.2.4

disable ...........................................................................................................................................

5-4

 

 

5.2.5

enable ............................................................................................................................................

5-4

 

 

5.2.6

get ..................................................................................................................................................

5-5

 

 

5.2.7

load ................................................................................................................................................

5-5

 

 

5.2.8

pause .............................................................................................................................................

5-5

 

 

5.2.9

resume...........................................................................................................................................

5-5

 

 

5.2.10

save ................................................................................................................................................

5-5

 

 

5.2.11

unload ...........................................................................................................................................

5-5

 

 

5.2.12

update ...........................................................................................................................................

5-6

 

5.3 Accessing Routing Service from a Connext Application.....................................................................

5-8

6

Monitoring Routing Service from a Remote Location

 

 

6.1

Enabling Remote Monitoring...................................................................................................................

6-2

 

6.2

Monitoring Configuration Data ..............................................................................................................

6-2

 

 

6.2.1

Configuration Data for Routing Service ..................................................................................

6-2

 

 

6.2.2

Configuration Data for a Domain Route..................................................................................

6-4

 

 

6.2.3

Configuration Data for a Session ..............................................................................................

6-5

 

 

6.2.4

Configuration Data for a Route.................................................................................................

6-6

 

 

6.2.5

Configuration Data for an Auto Route....................................................................................

6-11

 

6.3

Monitoring Status ....................................................................................................................................

6-15

 

 

6.3.1

How the Statistics are Generated ............................................................................................

6-16

 

 

6.3.2

Status Information for the Routing Service ...........................................................................

6-17

 

 

6.3.3

Domain Route Status ................................................................................................................

6-19

 

 

6.3.4

Status Information for a Session .............................................................................................

6-20

 

 

6.3.5

Status Information for a Route ................................................................................................

6-21

 

 

6.3.6

Status Information for an Auto Route ...................................................................................

6-22

7

Traversing Wide Area Networks

 

 

7.1

TCP Communication Scenarios...............................................................................................................

7-2

 

 

7.1.1

Communication Within a Single LAN .....................................................................................

7-2

 

 

7.1.2

Symmetric Communication Across NATs ...............................................................................

7-2

 

 

7.1.3

Asymmetric Communication Across NATs ............................................................................

7-3

 

 

7.1.4

Secure Communication ..............................................................................................................

7-3

 

7.2 Configuring the TCP Transport...............................................................................................................

7-4

 

 

7.2.1

TCP Transport Initial Peers ........................................................................................................

7-4

 

 

7.2.2

Setting Up the TCP Transport Properties with the PropertyQoSPolicy..............................

7-5

 

 

7.2.3

TCP/TLS Transport Properties .................................................................................................

7-7

8

Extending Routing Service with Adapters

 

 

8.1 Adapter Usage and Configuration..........................................................................................................

8-1

 

8.2 Adapter API And Entity Model ..............................................................................................................

8-3

 

 

8.2.1

Entity Creation.............................................................................................................................

8-8

 

 

8.2.2

Stream Discovery.........................................................................................................................

8-9

 

 

8.2.3

Reading Data..............................................................................................................................

8-10

iv

8.3 Creating New Adapters..........................................................................................................................

8-10

8.3.1

Adapter SDK Components ......................................................................................................

8-10

8.3.2

C Adapter API ............................................................................................................................

8-11

8.3.3 My First C Adapter ...................................................................................................................

8-13

8.3.4

Debugging C Adapters.............................................................................................................

8-33

8.3.5

Java Adapter API.......................................................................................................................

8-35

8.3.6 My First Java Adapter...............................................................................................................

8-37

8.3.7

Debugging Java Adapters ........................................................................................................

8-52

8.3.8

Testing an Adapter ....................................................................................................................

8-55

v

Chapter 1 Welcome to RTI Routing Service

Welcome to RTI® Routing Service, an out- of-the-box solution for integrating disparate and geographically dispersed systems. It scales RTI Connext™ (formerly RTI Data Distribution Service) applications across domains, LANs and WANs, including firewall and NAT traversal. Routing Service also supports Connext-to-Connext bridging by allowing you to make transformations in the data along the way. This allows unmodified Connext applications to communicate even if they were developed using incompatible interface definitions. This is often the case when integrating new and legacy applications or independently developed systems. Using RTI Routing Service Adapter SDK, you can extend Routing Service to interface with non-

Connext systems using off-the-shelf or custom developed adapters, including to third-party JMS implementations and legacy code written to the network socket API.

Traditionally, Connext applications can only communicate with applications in the same domain. With Routing Service, you can send and receive data across domains. You can even transform and filter the data along the way! Not only can you change the actual data values, you can change the data’s type. So the sending and receiving applications don’t even need to use the same data structure. You can also control which data is sent by using allow and deny lists.

Simply set up Routing Service to pass data from one domain to another and specify any desired data filtering and transformations. No changes are required in the Connext applications.

Key benefits of Routing Service:

It can significantly reduce the time and effort spent integrating and scaling Connext applications across Wide Area Networks and Systems-of-Systems.

Many systems today already rely on Connext to distribute their information across a Local Area Network (LAN). However, more and more of these systems are being integrated in Wide Area Networks (WANs). With Routing Service, you can scale Connext real-time publish/subscribe data-distribution beyond the current local networks and

1-1

make it available throughout a WAN—without making any changes to existing Connext applications. You can take an existing, even deployed system and integrate it with new applications or other existing systems without changing those existing systems.

With Routing Service, you can build modular systems out of existing systems. Data can be contained in private domains within subsystems and you can designate that only certain “global topics” can be seen across domains. The same mechanism controls the scope of discovery. Both application-level and discovery traffic can be scoped, facilitating scalable designs.

Routing Service provides secure deployment across multiple sites. You can partition networks and protect them with firewalls and NATS and precisely control the flow of data between the network segments.

It allows you to manage the evolution of your data model at the subsystem level. You can use Routing Service to transform data on the fly, changing topic names, type definitions, QoS, etc., seamlessly bridging different generations of Connext topic definitions.

Routing Service provides features for development, integration and testing. Multiple sites can each locally test and integrate their core application, expose selected topics of data, and accept data from remote sites to test integration connectivity, topic compatibility and specific use-cases.

It connects remotely to live, deployed systems so you can perform live data analytics, fault condition analysis, and data verification.

RTI Routing Service Adapter SDK allows you to quickly build and deploy bridges to integrate Connext and non-Connext systems. This can be done in a fraction of the time required to develop completely custom solutions. Bridges automatically inherit advanced Connext capabilities, including automatic discovery of applications; data transformation and filtering; data lifecycle management and support across operating systems; programming languages and network transports.

1-2

RTI Routing Service Adapter SDK offers an out-of-the-box solution for interfacing with third-party protocols and technology. It includes prebuilt adapters that can be used out-of-the-box to interface with third- party Java Message Service (JMS) providers or legacy code written to the network socket API. Adapters include source code so they can be easily modified to meet application- specific requirements or serve as a template for quick creation of new custom adapters.

Quickly build and deploy bridges between natively incompatible protocols and technologies using Connext

1.1Available Documentation

Routing Service documentation includes:

Getting Started Guide (RTI_Routing_Service_GettingStarted.pdf)—Highlights the benefits of Routing Service. It provides installation and startup instructions, and walks you through several examples so you can quickly see the benefits of using Routing Service.

Release Notes (RTI_Routing_Service_ReleaseNotes.pdf)—Describes system requirements and compatibility, as well as any version-specific changes and known issues.

User’s Manual (RTI_Routing_Service_UsersManual.pdf)—Describes how to configure Routing Service and use it remotely.

If the optional RTI Routing Service Adapter SDK is installed, you will also have the following documents:

RTI Routing Service Adapter SDK Installation Guide

(RTI_Routing_Service_AdapterSDK_InstallationGuide.pdf)—Describes installation instructions for RTI Routing Service Adapter SDK.

1-3

RTI Routing Service Adapter SDK Release Notes

(RTI_Routing_Service_AdapterSDK_ReleaseNotes.pdf)—Describes system requirements and compatibility, as well as any version-specific changes and known issues for RTI Routing Service Adapter SDK.

1-4

Chapter 2 Configuring Routing Service

This document describes how to configure Routing Service. To see installation instructions, or to walk through some simple examples, please see the Getting Started Guide.

When you start Routing Service, you can specify a configuration file in XML format (it is not required). In that file, you can set properties that control the behavior of the service. This chapter describes how to write a configuration file.

This chapter describes:

Terms to Know (Section 2.1)

How to Load the XML Configuration (Section 2.2)

XML Syntax and Validation (Section 2.3)

XML Tags for Configuring Routing Service (Section 2.4)

Enabling and Disabling Routing Service Entities (Section 2.5)

Enabling RTI Distributed Logger in Routing Service (Section 2.6)

Support for Extensible Types (Section 2.7)

2.1Terms to Know

Before learning how to configure Routing Service, you should become familiar with a few key terms and concepts.

A routing service entity refers to an execution of Routing Service.

A domain route defines a two-way mapping between two data domains. For example, a domain route could define a mapping between two different domains or between a Connext domain and a JMS provider's network.

A session defines a single-threaded context for routes. Data cannot be read and written from two routes in the session concurrently.

A route defines a one-way mapping between an “input” stream in one domain and an “output” stream in the other domain. For example, in a route between Connext and JMS, the input stream will be a Connext topic and the output stream will be a JMS topic or queue.

An auto route defines a set of potential routes that can be instantiated based on deny/ allow filters on the stream name and registered type name.

2-1

A transformation is a pluggable component that changes data from the “input” stream A to data in the “output” stream B.

An adapter is a pluggable component that allows Routing Service to consume and produce data for different data domains. By default, Routing Service is distributed with a built-in Connext adapter.

2.2How to Load the XML Configuration

Routing Service loads its XML configuration from multiple locations. This section presents the various approaches, listed in load order.

The first three locations only contain QoS Profiles and are inherited from Connext (see Chapter

15in the RTI Core Libraries and Utilities User's Manual).1

$NDDSHOME/resource/qos_profiles_5.0.x2/xml/NDDS_QOS_PROFILES.xml

This file contains the Connext default QoS values; it is loaded automatically if it exists.

(First to be loaded.)

File in NDDS_QOS_PROFILES

The files (or XML strings) separated by semicolons referenced in this environment vari- able are loaded automatically.

<working directory>/USER_QOS_PROFILES.xml

This file is loaded automatically if it exists.

The next locations are specific to Routing Service.

<Routing Service executable location>/../../resource/xml/ RTI_ROUTING_SERVICE.xml

This file contains the default Routing Service configuration; it is loaded if it exists. RTI_ROUTING_SERVICE.xml defines a service that automatically routes all types and topics between domains 0 and 1.

<working directory>/USER_ROUTING_SERVICE.xml

This file is loaded automatically if it exists.

File specified using the command line parameter -cfgFile

The command-line option -cfgFile (see Table 3.1 on page 3-2) can be used to specify a configuration file.

File specified using the remote command ‘load’

The load command (see Section 5.2.7) allows loading an XML file remotely. The file loaded using this command replaces to the file loaded using the -cfgFile command-line option. (Last to be loaded.)

You may use a combination of the above approaches.

Figure 2.1 shows an example configuration file. You will learn the meaning of each line as you read the rest of this chapter.

1.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2.x stands for the version number of the current release.

2-2

Figure 2.1 Example XML Configuration File

<?xml version="1.0"?> <dds>

<routing_service name="TopicBridgeExample" group_name="MyGroup"> <domain_route name="DomainRoute">

<participant_1> <domain_id>0</domain_id>

</participant_1>

<participant_2> <domain_id>1</domain_id>

</participant_2>

<session name="Session">

<topic_route name="SquaresToCircles"> <input participant="1">

<registered_type_name>ShapeType</registered_type_name> <topic_name>Square</topic_name>

</input>

<output> <registered_type_name>ShapeType</registered_type_name> <topic_name>Circle</topic_name>

</output> </topic_route>

</session> </domain_route>

</routing_service> </dds>

This file configures a simple bridge from Connext domain 0 to Connext domain 1 and changes the data’s topic from Square to Circle. Both topics use the same data type (ShapeType). You will find this example in <Routing Service installation directory>/example/shapes/topic_bridge.xml. Additional examples are in the same directory.

2.3XML Syntax and Validation

The XML configuration file must follow these syntax rules:

The syntax is XML; the character encoding is UTF-8.

Opening tags are enclosed in <>; closing tags are enclosed in </>.

A tag value is a UTF-8 encoded string. Legal values are alphanumeric characters. Routing Service’s parser will remove all leading and trailing spaces1 from the string before it is processed.

For example, " <tag> value </tag>" is the same as "<tag>value</tag>".

All values are case-sensitive unless otherwise stated.

Comments are enclosed as follows: <!-- comment -->.

The root tag of the configuration file must be <dds> and end with </dds>.

1.Leading and trailing spaces in enumeration fields will not be considered valid if you use the distributed XSD doc- ument to do validation at run-time with a code editor.

2-3

Routing Service provides DTD and XSD files that describe the format of the XML content. We recommend including a reference to one of these documents in the XML file that contains the routine service’s configuration—this provides helpful features in code editors such as Visual Studio and Eclipse, including validation and auto-completion while you are editing the XML file.

The DTD and XSD definitions of the XML elements are in <Routing Service installation directory>/resource/schema/rti_routing_service.dtd and <Routing Service installation directory>/resource/schema/rti_routing_service.xsd, respectively.

To include a reference to the XSD document in your XML file, use the attribute xsi:noNamespaceSchemaLocation in the <dds> tag. For example:

<?xml version="1.0" encoding="UTF-8"?>

<dds xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation=

"<Routing Service install directory>/resource/schema/rti_routing_service.xsd">

...

</dds>

To include a reference to the DTD document in your XML file, use the <!DOCTYPE> tag.

For example:

<?xml version="1.0" encoding="UTF-8"?>

<!DOCTYPE dds SYSTEM "<Routing Service install directory> /resource/schema/rti_routing_service.dtd">

<dds>

...

</dds>

We recommend including a reference to the XSD file in the XML documents; this provides stricter validation and better auto-completion than the corresponding DTD file.

2-4

2.4XML Tags for Configuring Routing Service

This section describes the XML tags you can use in a Routing Service configuration file. The following diagram and Table 2.1 describe the top-level tags allowed within the root <dds> tag.

See RTI Core Libraries and Utilities User’s Manual (Ch. 15) (see note below)

Section 2.4.6.2

Section 2.4.6.5

Section 2.4.8

Section 2.4.1

Note: The RTI Core Libraries and Utilities User’s Manual is located in <Connext installation directory>/ ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

Table 2.1 Top-level Tags in the Configuration File

 

 

Number

Tags within <dds>

Description

of Tags

 

 

Allowed

 

 

 

 

Specifies a library of adapter plugins.

 

<adapter_library>

See Adapters (Section 2.4.8) and Chapter 8: Extending Routing

0 or more

 

Service with Adapters.

 

 

 

 

 

Specifies a QoS library and profiles.

 

<qos_library>

The contents of this tag are specified in the same manner as for a

0 or more

 

Connext QoS profile file—see Chapter 15 in the RTI Core Libraries

 

 

and Utilities User’s Manual.1

 

<routing_service>

Specifies a Routing Service configuration. See Routing Service

1 or more

(Section 2.4.1).

(required)

 

 

 

 

2-5

Table 2.1 Top-level Tags in the Configuration File

 

 

Number

Tags within <dds>

Description

of Tags

 

 

Allowed

 

 

 

 

Specifies a library of transformation plugins.

 

<transformation_library>

See Data Transformation (Section 2.4.6.5) and Chapter 4:

0 or more

 

Transforming Data with Routing Service.

 

 

 

 

<types>

Defines types that can be used by the routing service.

0 or 1

See Defining Types in the Configuration File (Section 2.4.6.2).

 

 

 

 

 

1.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2.4.1Routing Service

A configuration file must have at least one <routing_service> tag; this tag is used to configure an execution of Routing Service. A configuration file may contain multiple <routing_service> tags.

When you start Routing Service, you can specify which <routing_service> tag to use to configure the service using the -cfgName command-line parameter.

For example:

<dds>

<routing_service name="Router1" group_name=”Group1”>

...

</routing_service>

<routing_service name="Router2" group_name=”Group1”>

...

</routing_service>

</dds>

Starting Routing Service with the following command will use the <routing_service> tag with the name Router1:

rtiroutingservice -cfgFile example.xml -cfgName Router1

Because a configuration file may contain multiple <routing_service> tags, one file can be used to configure multiple Routing Service executions.

A routing service may belong to a group of several routing services identified by a common group_name. This common name can be used to implement a specific policy when the communication happens between routing services of the same group.

2-6

For example, in the built-in Connext adapter, a participant will ignore other participants in the same group, as a way to avoid circular communication.

Section 2.4.3

Section 2.4.4

Section 2.4.4

Section 2.4.2

If the <routing_service> tag does not have a group_name attribute, Routing Service will use the

 

following

name:

RTI_RoutingService_<Host

Name>_<Process

ID>,

such

as

 

RTI_RoutingService_myhost_20024.

 

 

 

 

 

Table 2.2

describes

the tags allowed within a <routing_service> tag. Notice that the

 

<domain_route> tag is required.

 

 

 

 

Table 2.2 Routing Service Tags

 

 

 

 

 

 

 

 

 

 

 

 

 

Tags within

 

 

 

Number

 

Description

 

of Tags

 

<routing_service>

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Enables and configures remote administration. See Administration

 

 

 

<administration>

(Section 2.4.3) and Chapter 5: Administering Routing Service from a

0 or 1

 

 

 

 

Remote Location.

 

 

 

 

 

 

 

 

 

 

 

 

 

Contains a <documentation> tag that can be used to provide a routing

 

 

 

<annotation>

service description. This description will show up when you run

0 or 1

 

 

 

 

Routing Service without the -cfgName command-line option.

 

 

 

 

 

 

 

 

 

 

 

 

<domain_route>

Defines a mapping between two data domains. See Section 2.4.2.

1 or

more

 

(required)

 

 

 

 

 

 

 

 

 

 

 

 

<entity_monitoring>

Enables and configures remote monitoring for the routing_service

0 or 1

 

 

 

 

entity.

 

 

 

 

2-7

Table 2.2 Routing Service Tags

Tags within

 

Number

Description

of Tags

<routing_service>

 

Allowed

 

 

 

Configures the Java JVM used to load and run Java adapters such as

 

 

the JMS Adapter. For example:

 

 

<jvm>

 

 

<class_path>

 

 

<element>SocketAdapter.jar</element>

 

 

</class_path>

 

 

<options>

 

 

<element>-Xms32m</element>

 

 

<element>-Xmx128m</element>

 

<jvm>

</options>

0 or 1

 

</jvm>

 

 

The class path for the Java adapters can be set using either the

 

 

<class_path> tag or by setting the CLASSPATH environment variable.

 

 

Routing Service will always add <Routing Service executable loca-

 

 

tion>/../../class/rtiroutingservicesdk.jar and <Routing Service execut-

 

 

able location>/../../class/dds.jar at the end of the user defined class

 

 

path.

 

 

You can use the <options> tag to specify options for the JVM, such as

 

 

the initial and maximum Java heap sizes.

 

 

 

 

 

Enables and configures general remote monitoring. General

 

<monitoring>

monitoring settings are applicable to all the Routing Service entities

0 or 1

 

unless they are explicitly overridden. See Monitoring (Section 2.4.4).

 

2-8

2.4.2Domain Route

A domain route defines a mapping between two data domains. Data available in either of these data domains can be routed to the other one. For example, a domain route could define a mapping between two different Connext domains or between a Connext domain and a JMS provider's network. How this data is actually read and written is defined in specific routes.

A domain route creates two connections, known as connection_1 and connection_2. Each connection belongs to one of the two data domains.

For example:

<dds>

<routing_service name="Router1" group_name="Group1">

<domain_route name="DomainRoute1"> <connection_1 plugin_name=”...”>

. . .

</connection_1>

<connection_2 plugin_name=”...”>

...

</connection_2>

<session name="Session">

...

</session> </domain_route>

...

</routing_service> </dds>

The connection tags require the specification of the attribute plugin_name, which will be used to associate a connection with an adapter plugin defined within <adapter_library> (see Section 2.1).

2-9

For Connext domains, the connections are specified using the tags participant_1 and participant_2. Each tag has one associated DomainParticipant.

The following example routes information between two Connext domains.

<dds>

<routing_service name="Router1" group_name="Group1">

<domain_route name="DomainRoute1"> <participant_1>

<domain_id>54</domain_id>

...

</participant_1>

<participant_2> <domain_id>55</domain_id>

...

</participant_2>

<session name="Session">

...

</session> </domain_route>

...

</routing_service> </dds>

Configurations mixing connections and participants are allowed to provide communication between Connext domains and other data domains.

The following example routes information between a JMS provider network and a Connext domain.

<dds>

<routing_service name="Router1" group_name="Group1"> <domain_route name="DomainRoute1">

<connection_1 plugin_name=”adapter_library::jms”>

...

</connection_1>

<participant_2> <domain_id>55</domain_id>

...

</participant_2>

<session name="Session">

...

</session> </domain_route>

...

</routing_service> </dds>

2-10

Table 2.3 lists the tags allowed within a <domain_route> tag. Notice that most of these tags are required.

Table 2.4 lists the tags allowed within <connection_1> and <connection_2> tags.

Table 2.5 lists the tags allowed within <participant_1> and <participant_2> tags. Notice that the <domain_id> tag is required.

Table 2.3 Domain Route Tags

 

Tags within

 

 

 

Number of

 

 

Description

 

Tags

 

<domain_route>

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<connection_1>

 

Applicable to non-Connext domains.

 

1 (required)

 

 

Configures the first connection. See Table 2.4.

 

 

 

 

 

 

 

 

 

 

 

 

 

<connection_2>

 

Applicable to non-Connext domains.

 

1 (required)

 

 

Configures the second connection. See Table 2.4.

 

 

 

 

 

 

 

 

 

 

 

 

 

<entity_monitoring>

 

Enables and configures remote monitoring for the domain route. See

 

0 or 1

 

 

 

Monitoring (Section 2.4.4).

 

 

 

<participant_1>

 

Only applicable to Connext domains.

 

1 (required)

 

 

Configures the first participant. See Table 2.5.

 

 

 

 

 

 

 

 

 

 

 

 

 

<participant_2>

 

Only applicable to Connext domains.

 

1 (required)

 

 

Configures the second participant. See Table 2.5.

 

 

 

 

 

 

 

 

 

 

 

 

 

<session>

 

Defines a single-threaded context in which data is routed according

 

1 or more

 

 

to specified routes. See Session (Section 2.4.5).

 

(required)

 

 

 

 

 

 

 

 

 

 

Table 2.4 Connection Tags

 

 

 

 

 

 

 

 

 

 

 

Tags within

 

 

 

Number

 

 

Description

 

of Tags

 

<connection_1/2>

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

Registers a type name and associates it with a type representation.

 

 

 

<registered_type>

When you define a type in the configuration file (with the <types> tag),

 

0 or more

 

 

you have to register the type in order to use it in routes. See Route

 

 

 

 

Types (Section 2.4.6.1).

 

 

 

 

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure the

 

 

 

 

parameters of the connection. For example:

 

 

 

 

 

<property>

 

 

 

 

 

<value>

 

 

 

<property>

 

<element>

 

0 or 1

 

 

<name>jms.connection.username</name>

 

 

 

 

 

 

 

 

 

<value>myusername</value>

 

 

 

 

 

</element>

 

 

 

 

 

</value>

 

 

 

 

 

</property>

 

 

 

 

 

 

 

 

2-11

Table 2.5 Participant Tags

Tags within

 

 

Number of

Description

 

Tags

<participant_1/2>

 

 

 

Allowed

 

 

 

 

 

 

 

<domain_id>

Sets the domain ID associated with the participant.

 

1 (required)

 

 

 

 

Configures certain aspects of how Connext allocates internal memory.

 

 

The configuration is per domain_route's participant and therefore

 

 

affects all the contained Connext readers and Connext writers. For

 

 

example:

 

 

 

<domain_route name="test">

 

 

 

<participant_1>

 

 

 

<domain_id>0</domain_id>

 

 

 

...

 

 

 

<memory_management>

 

 

 

<sample_buffer_min_size>X</sample_buffer_min_size>

 

 

 

<sample_buffer_trim_to_size>

 

 

 

true

 

 

 

</sample_buffer_trim_to_size>

 

 

 

</memory_management>

 

 

 

</participant_1>

 

 

<memory_

...

 

 

 

 

0 or more

management>

The <memory_management> tag can include the following tags:

 

 

 

 

sample_buffer_min_size: For all Connext readers/writers,

the

 

 

way Connext allocates memory for samples is as follows: Connext

 

 

pre-allocates space for samples up to size X in the reader and

 

 

writer queues. If a sample has an actual size greater than X, the

 

 

memory is allocated dynamically for that sample. The default size

 

 

is DDS_LENGTH_UNLIMITED (meaning no dynamic memory is

 

 

used; the maximum sample size is pre-allocated).

 

 

 

sample_buffer_trim_to_size: If set to true, after allocating

 

 

dynamic memory for very large samples, that memory will be

 

 

released when possible. If false, that memory will not be released

 

 

but kept for future samples if needed. The default is false.

 

 

 

This feature is useful when a data type has a very high maximum size

 

 

(e.g., megabytes) but most of the samples sent are much smaller than

 

 

the maximum possible size (e.g., kilobytes). In this case, the memory

 

 

footprint is reduced dramatically, while still correctly handling the

 

 

rare cases in which very large samples are published.

 

 

 

 

 

 

2-12

Table 2.5 Participant Tags

Tags within

 

Number of

Description

Tags

<participant_1/2>

 

Allowed

 

 

 

 

 

 

Registers a type name and associates it with a type code. When you

 

<registered_type>

define a type in the configuration file (with the <types> tag), you have

0 or more

 

to register the type in order to use it in topic routes. See Route Types

 

 

(Section 2.4.6.1).

 

 

 

 

 

Sets the participant QoS.

 

 

The contents of this tag are specified in the same manner as a Connext

 

 

QoS profile file—see Chapter 15 in the RTI Core Libraries and Utilities

 

 

User’s Manual.1

 

 

If not specified, the default is used.

 

 

You can use a <participant_qos> tag inside a <qos_library>/

 

 

<qos_profile> previously defined in your configuration file by

 

 

referring to it like this:

 

 

<participant_qos base_name="MyLibrary::MyProfile"/>

 

<participant_qos>

To use that profile but override just some values:

0 or 1

<participant_qos base_name="MyLibrary::MyProfile">

 

 

 

<discovery>

 

 

<initial_peers>

 

 

<element>udpv4://192.168.1..12</element>

 

 

<element>shmem://</element>

 

 

</initial_peers>

 

 

</discovery>

 

 

</participant_qos>

 

 

(This applies to all QoS tags: <publisher_qos>, <subscriber_qos> in

 

 

sessions; <datareader_qos>, <datawriter_qos> in topic routes and

 

 

auto topic routes.)

 

 

 

 

1.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2.4.3Administration

You can create a Connext application that can remotely control Routing Service. The <administration> tag is used to enable remote administration and configure its behavior.

By default, remote administration is turned off in Routing Service for security reasons. A remote administration section is not required in the configuration file.

For example:

<dds>

<routing_service>

<administration>

<domain_id>55</domain_id>

<save_path>

/home/david/mysaved_config.xml

</save_path>

</administration>

...

</routing_service>

</dds>

When remote administration is enabled, Routing Service will create a DomainParticipant, Publisher, Subscriber, DataWriter, and DataReader. These entities are used to receive commands and send responses. You can configure these entities with QoS tags within the <administration> tag.

2-13

Table 2.6 lists the tags allowed within <administration> tag. Notice that the <domain_id> tag is required.

For more details, please see Chapter 5: Administering Routing Service from a Remote Location.

Note: The command-line options used to configure remote administration take precedence over the XML configuration (see Table 3.1 on page 3-2).

Table 2.6 Remote Administration Tags

Tags within

 

Number

Description

of Tags

<administration>

 

Allowed

 

 

 

 

 

 

 

 

 

A boolean that, if true, automatically triggers a save command when

 

 

configuration updates are received. It is false by default.

 

<autosave_on_

This value is mutable when an update (Section 5.2.12) command

 

targets a routing service.

0 or 1

update>

This value is sent as part of the monitoring configuration data for the

 

 

 

 

routing service (see Configuration Data for Routing Service (Section

 

 

6.2.1)).

 

 

 

 

 

Configures the DataReader QoS for remote administration.

 

 

If the tag is not defined, Routing Service will use the Connext defaults

 

 

with the following changes:

 

<datareader_qos>

reliability.kind = DDS_RELIABLE_RELIABILITY_QOS (this value

0 or 1

 

cannot be changed)

 

 

history.kind = DDS_KEEP_ALL_HISTORY_QOS

 

 

resource_limits.max_samples = 32

 

 

 

 

 

Configures the DataWriter QoS for remote administration.

 

 

If the tag is not defined, Routing Service will use the Connext defaults

 

<datawriter_qos>

with the following changes:

0 or 1

 

history.kind = DDS_KEEP_ALL_HISTORY_QOS

 

 

resource_limits.max_samples = 32

 

 

 

 

<distributed_logger>

Configures RTI Distributed Logger.

0 or 1

See Enabling RTI Distributed Logger in Routing Service (Section 2.6).

 

 

 

 

 

<domain_id>

Specifies which domain ID Routing Service will use to enable remote

1

administration.

(required)

 

 

 

 

<participant_qos>

Configures the DomainParticipant QoS for remote administration.

0 or 1

If the tag is not defined, Routing Service will use the Connext defaults.

 

 

 

 

 

<publisher_qos>

Configures the Publisher QoS for remote administration.

0 or 1

If the tag is not defined, Routing Service will use the Connext defaults.

 

 

 

 

 

 

Specifies the file that will contain the saved configuration. It is empty

 

 

by default.

 

 

A <save_path> must be specified if you want to use the save (Section

 

 

5.2.10) command. If the file specified by <save_path> already exists,

 

<save_path>

the file will be overwritten when save is executed.

0 or 1

This value is mutable when an update (Section 5.2.12) command

 

 

 

targets a routing service.

 

 

This value is sent as part of the monitoring configuration data for the

 

 

routing service (see Configuration Data for Routing Service (Section

 

 

6.2.1)).

 

 

 

 

<subscriber_qos>

Configures the Subscriber QoS for remote administration.

0 or 1

If the tag is not defined, Routing Service will use the Connext defaults.

 

 

 

 

 

2-14

2.4.4Monitoring

You can create a Connext application that can remotely monitor the status of Routing Service. To enable remote monitoring and configure its behavior, use the <monitoring> and

<entity_monitoring> tags.

By default, remote monitoring is turned off in Routing Service for security and performance reasons. A remote monitoring section is not required in the configuration file.

For example:

<dds>

<routing_service>

<enabled>true</enabled>

<monitoring>

<domain_id>55</domain_id>

<status_publication_period>

<sec>1</sec>

</status_publication_period>

</monitoring>

...

</routing_service>

</dds>

Routing Service allows monitoring of the following kinds of entities:

<routing_service> (see Section 2.4.1)

<domain_route> (see Section 2.4.2)

<session> (see Section 2.4.5)

<route> (see Section 2.4.6)

<topic_route> (see Section 2.4.6)

<auto_route> (see Section 2.4.7)

<auto_topic_route> (see Section 2.4.7)

For each entity, Routing Service can publish two kinds of information:

Entity data

Entity status

Entity data provides information about the configuration of the entity. For example, the route data contains information such as the stream name and the type name. Entity data information is republished every time the entity is enabled, disabled or has configuration changes.

Entity status provides information about the operational status of an entity. This kind of information changes continuously and is computed and published periodically. For example, the route status contains information such as the route’s latency and throughput.

For more information about entity data and status, see Chapter 6: Monitoring Routing Service from a Remote Location.

When remote monitoring is enabled, Routing Service will create one DomainParticipant, one Publisher, five DataWriters for data publication (one for each kind of entity), and five DataWriters for status publication (one for each kind of entity). You can configure the QoS of these entities with the <monitoring> tag defined under <routing_service>.

The general remote monitoring parameters specified using the <monitoring> tag in

<routing_service> (except domain_id, participant_qos, publisher_qos, and datawriter_qos) can be overwritten on a per entity basis using the <entity_monitoring> tag.

For example:

<dds> <routing_service>

2-15

<monitoring>

<domain_id>55</domain_id> <status_publication_period>

<sec>1</sec>

</status_publication_period>

</monitoring>

...

<domain_route>

<entity_monitoring>

<status_publication_period> <sec>4</sec>

</status_publication_period> </entity_monitoring>

...

</domain_route> </routing_service>

</dds>

Table 2.7 lists the tags allowed within <monitoring> tag.

Table 2.7 Monitoring tags

Tags within

 

Number of

Description

Tags

<monitoring>

 

Allowed

 

 

 

 

 

 

Configures the DataWriter QoS for remote monitoring.

 

<datawriter_qos>

If the tag is not defined, Routing Service will use the Connext defaults

0 or 1

 

with the following change:

 

 

durability.kind = DDS_TRANSIENT_LOCAL_DURABILITY_QOS

 

 

 

 

<domain_id>

Specifies which domain ID Routing Service will use to enable remote

1 (required)

 

monitoring.

 

 

Enables/disables general remote monitoring.

 

 

Setting this value to true (default value) in the <monitoring> tag

 

 

under <routing_service> enables monitoring in all the entities

 

 

unless they explicitly disable it by setting this tag to false in their

 

<enabled>

local <entity_monitoring> tags.

0 or 1

 

Setting this tag to false in the <monitoring> tag under

 

 

<routing_service> disables monitoring in all the Routing Service

 

 

entities. In this case, any monitoring configuration settings in the

 

 

entities are ignored.

 

 

 

 

2-16

Table 2.7 Monitoring tags

Tags within

 

Number of

Description

Tags

<monitoring>

 

Allowed

 

 

 

 

 

 

Enables or disables the publication of statistics calculated within

 

 

fixed time windows.

 

 

By default, Routing Service only publishes the statistics

 

 

corresponding to the window between two status publications.

 

 

By using this tag, you can get the following additional windows:

 

 

5 seconds

 

 

1 minute

 

 

5 minutes

 

 

1 hour

 

<historical_statistics>

Up time (since the entity was enabled)

0 or 1

For example:

 

 

 

<historical_statistics>

 

 

<five_second>true</five_second>

 

 

<one_minute>true</one_minute>

 

 

<five_minute>false</five_minute>

 

 

<one_hour>true</one_hour>

 

 

<up_time>false</up_time>

 

 

</historical_statistics>

 

 

If a window is not present (inside the tag <historical_statistics>), it

 

 

is considered disabled.

 

 

Historical statistics can be overwritten on a per entity basis.

 

 

 

 

 

Configures the DomainParticipant QoS for remote monitoring.

 

<participant_qos>

If the tag is not defined, Routing Service will use the Connext defaults

0 or 1

 

with the following change:

 

 

resource_limits.type_code_max_serialized_length = 4096

 

 

 

 

 

Configures the Publisher QoS for remote monitoring.

 

<publisher_qos>

If the tag is not defined, Routing Service will use the Connext

0 or 1

 

defaults.

 

 

 

 

 

Specifies the frequency at which status statistics are gathered.

 

 

Statistical variables such as latency, are part of the entity status. For

 

 

example:

 

 

<statistics_sampling_period>

 

 

<sec>1</sec>

 

 

<nanosec>0</nanosec>

 

<statistics_sampling_

</statistics_sampling_period>

0 or 1

period>

 

The statistics period for a given entity should be smaller than the

 

 

 

 

publication period.

 

 

If the tag is not defined, the period is 1 second.

 

 

The statistics sampling period defined in <routing_service> is

 

 

inherited by all the entities inside <routing_service>.

 

 

An entity can overwrite the period.

 

 

 

 

2-17

Table 2.7 Monitoring tags

Tags within

 

Number of

Description

Tags

<monitoring>

 

Allowed

 

 

 

 

 

 

Specifies the frequency at which the status of an entity is published.

 

 

For example:

 

 

<status_publication_period>

 

 

<sec>3</sec>

 

<status_publication_

<nanosec>0</nanosec>

0 or 1

</status_publication_period>

period>

 

 

If the tag is not defined, the period is 5 seconds.

 

 

The status publication period defined in <routing_service> is

 

 

inherited by all the entities inside <routing_service>.

 

 

An entity can overwrite the period.

 

 

 

 

2.4.4.1Monitoring Configuration Inheritance

The monitoring configuration defined in <routing_service> is inherited by all the entities defined inside the tag.

An entity can overwrite three elements of the monitoring configuration:

The status publication period

The statistics sampling period

The historical statistics windows

Each one of this three elements is inherited and can be overwritten independently using the

<entity_monitoring> tag.

For example:

<dds>

<routing_service name=”MonitoringExample”> <monitoring>

<domain_id>55</domain_id> <status_publication_period>

<sec>1</sec> </status_publication_period> <statistics_sampling_period>

<sec>1</sec>

<nanosec>0</nanosec> </statistics_sampling_period>

</monitoring>

...

<domain_route> <entity_monitoring>

<status_publication_period> <sec>4</sec>

</status_publication_period> </entity_monitoring>

...

</domain_route> </routing_service>

</dds>

In the previous example, the domain route overwrites the status publication period to 4 seconds and inherits the statistics sampling period.

2-18

Table 2.8 Entity Monitoring Tags

Tags within

 

 

 

 

Number

 

Description

 

of Tags

<entity_monitoring>

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

Enables/disables remote monitoring for a given entity.

 

 

<enabled>

If general monitoring is disabled this value is ignored.

 

0 or 1

 

Default value: true

 

 

 

 

 

 

 

 

Enables or disables the publication of statistics calculated within fixed

 

 

time windows.

 

 

 

 

 

By default, Routing Service only publishes the statistics corresponding

 

 

to the window between two status publications.

 

 

 

By using this tag, you can get the following additional windows:

 

 

 

5 seconds

 

 

 

 

 

1 minute

 

 

 

 

 

5 minutes

 

 

 

 

 

1 hour

 

 

 

 

 

Up time (since the entity was enabled)

 

 

<historical_statistics>

For example:

 

 

 

0 or 1

 

<historical_statistics>

 

 

 

 

<five_second>true</five_second>

 

 

 

 

<one_minute>true</one_minute>

 

 

 

 

<five_minute>false</five_minute>

 

 

 

 

<one_hour>true</one_hour>

 

 

 

 

<up_time>false</up_time>

 

 

 

 

</historical_statistics>

 

 

 

 

If a window is not present (inside the tag <historical_statistics>), it is

 

 

considered disabled.

 

 

 

 

 

If this tag is not defined, historical statistics are inherited from the

 

 

general monitoring settings.

 

 

 

 

 

 

 

Specifies the frequency at which status statistics are gathered. Statistical

 

 

variables such as latency, are part of the entity status. For example:

 

 

 

<statistics_sampling_period>

 

 

 

 

<sec>1</sec>

 

 

 

 

 

<nanosec>0</nanosec>

 

 

 

<statistics_sampling_

</statistics_sampling_period>

 

 

 

The statistics period for a given entity should be smaller than the

0 or 1

period>

 

publication period.

 

 

 

 

 

If the tag is not defined, the period is inherited from the general

 

 

monitoring settings.

 

 

 

 

 

This tag is only present in the <entity_monitoring> tag of <route>,

 

 

<topic_route>,

<auto_route>,

<auto_topic_route>

and

 

 

<routing_service>.

 

 

 

 

 

 

 

 

Specifies the frequency at which the status of an entity is published. For

 

 

example:

 

 

 

 

 

<status_publication_period>

 

 

 

<status_publication_

<sec>3</sec>

 

 

 

0 or 1

period>

<nanosec>0</nanosec>

 

 

 

 

 

 

</status_publication_period>

 

 

 

 

If the tag is not defined, its value is inherited from the general

 

 

monitoring settings.

 

 

 

 

 

 

 

 

 

 

2-19

2.4.5Session

A <session> tag defines a single-threaded context for data routing; The data is routed according

to specified routes (Section 2.4.6) and auto routes (Section 2.4.7).

Each session will have an associated session thread that will serialize access to the routes in the session.

For example:

<dds>

...

<routing_service name=”MyRoutingService”>

...

<domain_route>

...

<session name="Session1">

...

<route name=”Route1” >

...

</route>

...

</session>

...

</domain_route>

...

</routing_service>

...

</dds>

Sessions that bridge Connext domains will create a Publisher and a Subscriber in the participants (participant_1 or participant_2) associated with the Connext domains.

Table 2.9 lists the tags allowed within a <session> tag.

Table 2.9 Session Tags

Tags within

 

Number

Description

of Tags

<session>

 

Allowed

 

 

 

 

 

 

 

 

<auto_route>

Defines a general route based on type and stream filters. See Auto Routes

0 or more

 

(Section 2.4.7).

 

<auto_topic_

Defines a general Connext topic route based on type and topic filters. See Auto

0 or more

route>

Routes (Section 2.4.7).

 

 

Enables and configures remote monitoring for the session. See Monitoring

 

<monitoring>

(Section 2.4.4) and Chapter 6: Monitoring Routing Service from a Remote

0 or 1

 

Location.

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure certain

 

 

parameters of the session. For example:

 

 

<property>

 

 

<value>

 

 

<element>

 

<property>

<name>com.rti.socket.timeout</name>

0 or 1

 

<value>1</value>

 

 

</element>

 

 

</value>

 

 

</property>

 

 

These properties are only used in non-Connext domains.

 

 

 

 

2-20

Table 2.9 Session Tags

Tags within

 

Number

Description

of Tags

<session>

 

Allowed

 

 

 

 

 

 

Only applicable to Connext.

 

 

Sets the QoS associated with the session Publishers. There is one Publisher per

 

<publisher_

participant.

 

The contents of this tag are specified in the same manner as a Connext QoS

0 or 1

qos>

profile file—see the chapter Configuring QoS with XML in the RTI Core Libraries

 

 

 

 

and Utilities User’s Manual.1

 

 

If the tag is not defined, Routing Service will use the Connext defaults.

 

 

 

 

<route>

Defines a data mapping between two streams. See Routes (Section 2.4.6)

0 or more

 

 

 

 

Only applicable to Connext.

 

 

Sets the QoS associated with the session Subscribers. There is one Subscriber

 

<subscriber_

per participant.

 

The contents of this tag are specified in the same manner as a Connext QoS

0 or 1

qos>

profile file—see the chapter Configuring QoS with XML in the RTI Core Libraries

 

 

 

 

and Utilities User’s Manual.1

 

 

If the tag is not defined, Routing Service will use the Connext defaults.

 

 

 

 

 

Sets the mask, priority and stack size of the thread associated with this session.

 

 

Example:

 

 

<session>

 

 

<thread>

 

 

<mask>MASK_DEFAULT</mask>

 

 

<priority>THREAD_PRIORITY_DEFAULT</priority>

 

 

<stack_size>THREAD_STACK_SIZE_DEFAULT</stack_size>

 

<thread>

</thread>

0 or 1

 

...

 

 

</session>

 

 

Default values:

 

 

mask = MASK_DEFAULT

 

 

priority = THREAD_PRIORITY_DEFAULT

 

 

stack_size = THREAD_STACK_SIZE_DEFAULT

 

 

 

 

<topic_route>

Defines a data mapping between two Connext topics. See Routes (Section

0 or more

2.4.6).

 

 

 

2-21

Table 2.9 Session Tags

Tags within

 

Number

Description

of Tags

<session>

 

Allowed

 

 

 

 

 

 

Configures the WaitSet used to sleep and notify the session thread when data

 

 

is available.

 

 

Example:

 

 

<session>

 

 

<wait_set>

 

 

<max_event_count>5</max_event_count>

 

 

<max_event_delay>

 

 

<sec>1</sec>

 

 

<nanosec>0</nanosec>

 

 

</max_event_delay>

 

<wait_set>

</wait_set>

0 or 1

 

...

 

 

</session>

 

 

In the previous example, the session thread wakes up and tries to read data

 

 

after a 1 second timeout expires (max_event_delay) or after it has been notified

 

 

five times across routes that new data is available (max_event_count).

 

 

Default values:

 

 

max_event_count = 1

 

 

max_event_delay.sec = DURATION_INFINITE_SEC

 

 

max_event_delay.nanosec = DURATION_INFINITE_NSEC

 

 

 

 

1.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2.4.6Routes

A route explicitly defines a mapping between an “input” data stream on one domain and an “output” data stream on the other domain.

For example, the following route defines a mapping between a Connext topic called Square and a JMS queue called Square.

<dds>

...

<routing_service>

...

<domain_route> <participant_1>

<domain_id>54</domain_id> </participant_1>

<connection_2 plugin_name=”my_adapter_library::jms> </connection_2>

...

<session name="Session1">

...

<route name="DDSSquaresToJMSSquares">

<dds_input participant=”1”>

<topic_name>Square</topic_name>

<registered_type_name>ShapeType</registered_type_name>

...

</dds_input>

<output>

<stream_name>Square</topic_name>

<registered_type_name>ShapeType</registered_type_name>

2-22

...

</output>

...

</route>

</session>

...

</domain_route>

...

</routing_service>

...

</dds>

Connext inputs and outputs within a route are defined using the XML tags <dds_input> and <dds_output>. Input and outputs from other data domains are defined using the tags <input> and <output>. A topic route is a special kind of route that defines a mapping between an “input” topic on one Connext domain and an “output” topic on other Connext domain. For example, the following topic route will subscribe to topic Square on domain 54 and will republish those samples on domain 55 as samples of topic Circle.

<dds>

...

<routing_service>

...

<domain_route> <participant_1>

<domain_id>54</domain_id> </participant_1>

<participant_2> <domain_id>55</domain_id>

</participant_2>

...

<session name="Session1">

...

<topic_route name="SquaresToCircles">

<input participant=”1”>

<topic_name>Square</topic_name> <registered_type_name>ShapeType</registered_type_name>

...

</input>

<output>

<topic_name>Circle</topic_name>

<registered_type_name>ShapeType</registered_type_name>

...

</output>

...

</topic_route>

</session>

...

</domain_route>

...

</routing_service>

...

</dds>

In the previous example, the direction of the mapping is defined by the attribute participant of the tag <input>. Therefore, to change the above example to read Squares from domain 55 and

2-23

write Circles on domain 54, we would use <input participant=”2”>. There is an equivalent attribute for non-Connext inputs called connection.

Inputs and outputs in a route or topic route have an associated StreamReader and StreamWriter, respectively. For Connext domains, the StreamReader will contain a DataReader and the StreamWriter will contain a DataWriter. The Connext DataReaders and DataWriters belong to the corresponding session’s Subscriber and Publisher.

The read and write operations in a route will be performed in the context of the thread associated with the session.

Routes vs. Auto Routes: A route is an explicit route of data for two specific streams. An auto route (defined with a different tag, <auto_route>) is a way to automatically create routes based on filters—see Auto Routes (Section 2.4.7).

Table 2.10 lists the tags allowed within a <route>.

Table 2.11 lists the tags allowed within a <topic_route>.

Table 2.12 lists the tags allowed within the input and output tags in a <route> tag.

Table 2.13 lists the tags allowed within the Connext input and output tags. in a <route> or

<topic_route> tag.

Table 2.10 Route Tags

 

 

Number

Tags within <route>

Description

of Tags

 

 

Allowed

 

 

 

<dds_input>

Only applicable to Connext inputs.

1

Defines the route’s input topic. See Table 2.13.

(required)

 

 

 

 

<dds_output>

Only applicable to Connext outputs.

1

Defines the route’s output topic. See Table 2.13.

(required)

 

 

 

 

 

Configures remote monitoring for the route. See Monitoring (Section

 

<entity_monitoring>

2.4.4) and Chapter 6: Monitoring Routing Service from a Remote

0 or 1

 

Location.

 

 

 

 

<input>

Only applicable to non-Connext inputs.

1

Defines the route’s input stream. See Table 2.13.

(required)

 

 

 

 

<output>

Only applicable to non-Connext outputs.

1

Defines the route’s output stream. See Table 2.13.

(required)

 

 

 

 

 

When this tag is true, the data samples read from the input stream are

 

 

written into the output stream with the same timestamp that was

 

<publish_with_

associated with them when they were made available in the input

 

domain.

0 or 1

original_timestamp>

 

This option may not be applicable in some adapter implementations in

 

 

which the concept of timestamp is unsupported.

 

 

Default: false

 

 

 

 

 

Defines if the input connection will use types discovered in the output

 

 

connection and vice versa for the creation of StreamWriters and

 

<route_types>

StreamReaders in the route.

0 or 1

 

 

See Discovering Types (Section 2.4.6.3).

 

 

Default: false

 

 

 

 

<transformation>

Sets a data transformation to be applied for every data sample (see

0 or 1

 

Data Transformation (Section 2.4.6.5)).

 

2-24

Table 2.11 Topic Route Tags

 

Tags within

 

 

Number

 

Description

 

of Tags

 

<topic_route>

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

Configures remote monitoring for the topic route. See Monitoring

 

 

 

<entity_monitoring>

(Section 2.4.4) and Chapter 6: Monitoring Routing Service from a

 

0 or 1

 

 

Remote Location.

 

 

 

 

 

 

 

 

<input>

Defines the topic route’s input topic. See Table 2.13.

 

1 (required)

 

 

 

 

 

 

<output>

Defines the topic route’s output topic. See Table 2.13.

 

1 (required)

 

 

 

 

 

 

 

Indicates whether or not disposed samples (NOT_ALIVE_DISPOSE)

 

 

 

 

must be propagated by the topic route.

 

 

 

<propagate_dispose>

This action maybe be overwritten by the execution of a

 

0 or 1

 

 

transformation.

 

 

 

 

Default: true

 

 

 

 

 

 

 

 

 

Indicates whether or not NOT_ALIVE_NO_WRITERS samples must

 

 

 

 

be propagated by the topic route by using the unregister_instance()

 

 

 

<propagate_

operation

 

0 or 1

 

 

 

 

unregister>

This action maybe be overwritten by the execution of a

 

 

 

transformation.

 

 

 

 

Default: true

 

 

 

 

 

 

 

 

 

Writes the data sample as if they came from its original writer. Setting

 

 

 

<publish_with_

this option to true allows having redundant routing services and

 

0 or 1

 

original_info>

prevents the applications from receiving duplicate samples.

 

 

 

 

 

 

Default: false

 

 

 

 

 

 

 

 

<publish_with_

When this tag is set to true, the data samples are written with their

 

 

 

original source timestamp.

 

0 or 1

 

original_timestamp>

 

 

Default: false

 

 

 

 

 

 

 

 

 

 

 

 

 

Defines if the input domain will use types discovered in the output

 

 

 

 

domain and vice versa for the creation of DataWriters and

 

 

 

<route_types>

DataReaders in the topic route.

 

0 or 1

 

 

 

 

 

See Discovering Types (Section 2.4.6.3).

 

 

 

 

Default: false

 

 

 

 

 

 

 

 

<transformation>

Sets a data transformation to be applied for every data sample (see

 

0 or 1

 

 

Data Transformation (Section 2.4.6.5)).

 

 

Table 2.12 Input and Output Tags for a Route

 

 

 

 

 

 

 

 

Tags within

 

 

Number

 

<input> and

Description

 

of Tags

 

<output>

 

 

Allowed

 

 

 

 

 

 

 

Specifies when to create the StreamReader/StreamWriter.

 

 

 

<creation_mode>

Default: IMMEDIATE

 

0 or 1

 

See Creation Modes—Controlling when StreamReaders and

 

 

 

 

 

 

 

StreamWriters are Created (Section 2.4.6.4).

 

 

 

 

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure

 

 

 

 

certain parameters of the StreamReaders/StreamWriters. For example:

 

 

 

 

<property>

 

 

 

 

<value>

 

 

 

<property>

<element>

 

0 or 1

 

<name>com.rti.socket.port</name>

 

 

 

 

 

 

 

<value>16556</value>

 

 

 

 

</element>

 

 

 

 

</value>

 

 

 

 

</property>

 

 

 

 

 

 

 

2-25

Table 2.12 Input and Output Tags for a Route

 

Tags within

 

 

 

 

 

 

 

 

Number

 

<input> and

 

 

 

Description

 

 

 

 

of Tags

 

<output>

 

 

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

<registered_type_

 

Sets the registered type name of this stream. See Route Types (Section

 

1

 

name>

2.4.6.1).

 

 

 

 

(required)

 

 

 

 

 

 

 

 

 

 

 

 

<stream_name>

 

Sets the stream name.

 

 

 

 

1

 

 

 

 

 

 

(required)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2.13 Connext Input and Output Tags for a Route or Topic Route

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tags within

 

 

Tags within <

 

 

 

 

 

Number

 

<topic_route><input>

 

topic_route><output>

 

 

 

 

 

 

 

 

Description

 

 

 

of Tags

 

and

 

 

and

 

 

 

 

 

 

 

 

 

 

 

 

Allowed

 

<route><dds_input>

 

<route><dds_output>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<registered_type_name>

Sets the registered type name of this topic. See

 

1

 

Route Types (Section 2.4.6.1).

 

 

 

(required)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<topic_name>

 

Sets the topic name.

 

 

 

1

 

 

 

 

 

(required)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Specifies when to create the DataReader/

 

 

 

 

 

 

 

DataWriter.

 

 

 

 

 

 

<creation_mode>

Default: IMMEDIATE

 

 

 

0 or 1

 

See Creation

Modes—Controlling

when

 

 

 

 

 

 

 

 

 

 

 

 

 

StreamReaders

and StreamWriters

are

 

 

 

 

 

 

 

Created (Section 2.4.6.4).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Defines a SQL content filter for the

 

 

 

 

 

 

 

DataReader.

 

 

 

 

 

 

 

 

 

 

Example:

 

 

 

 

 

 

 

 

 

 

<topic_route>

 

 

 

 

 

 

 

 

 

 

...

 

 

 

 

 

 

 

 

 

 

<input>

 

 

 

 

 

 

 

 

 

 

...

 

 

 

 

 

 

<content_filter>

 

 

N/A

<content_filter>

 

 

 

0 or 1

 

 

 

<expression>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

x > 100

 

 

 

 

 

 

 

 

 

</expression>

 

 

 

 

 

 

 

 

 

</content_filter>

 

 

 

 

 

 

 

 

 

...

 

 

 

 

 

 

 

 

 

 

</input>

 

 

 

 

 

 

 

 

 

 

...

 

 

 

 

 

 

 

 

 

 

</topic_route>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sets the DataReader or DataWriter QoS.

 

 

 

 

 

 

 

 

 

The contents of this tag are specified in the

 

 

 

 

 

 

 

same manner as a Connext QoS profile file—

 

 

 

<datareader_qos>

 

 

<datawriter_qos>

see the chapter on Configuring QoS with XML

 

0 or 1

 

 

 

 

 

in the RTI Core Libraries and Utilities User’s

 

 

 

 

 

 

 

Manual.1

 

 

 

 

 

 

 

 

 

 

If the tag is not defined, Routing Service will

 

 

 

 

 

 

 

use the Connext defaults.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2-26

2.4.6.1Route Types

The tag <registered_type_name> within the <input> and <output> tags contains the registered type name of the stream. The actual definition of that type can be set in the configuration file (see Section 2.4.6.2) or it can be discovered by the connections (see Section 2.4.6.3).

2.4.6.2Defining Types in the Configuration File

To define and use a type in your XML configuration file:

1.Define your type within the <types> tag. (This is one of the top-level tags, see Table 2.1.)

2.Register it in the connection(s)/participant(s) where you will use it.

3.Refer to it in the domain route(s) that will use it.

For example:

<dds>

...

<types>

<struct name="PointType">

...

</struct>

</types>

...

<routing_service name=”MyRoutingService”>

...

<domain_route>

<connection_1>

...

<registered_type name="Position" type_name="PointType"/>

</connection_1>

<participant_2>

...

<registered_type name="Position" type_name="PointType"/>

</participant_2>

...

<session> <topic_route>

<input participant="2"> <registered_type_name>Position</registered_type_name>

</input>

<output>

...

</output> </topic_route> </session>

...

</domain_route>

...

</routing_service>

...

<dds>

2-27

The type description is done using the Connext XML format for type definitions. For more information, see Section 3.4 in the RTI Core Libraries and Utilities User's Manual.1

2.4.6.3Discovering Types

If a route refers to types that are not defined in the configuration file, Routing Service has to discover their type representation (e.g. typecode). A route cannot be created without the type representation information.

By default, the StreamReader creation will be tied to the discovery of types (e.g. typecodes) in the input domain and the StreamWriter creation will be tied to the discovery of types (e.g typecodes) in the output domain. If you want to use types discovered in either one of the domains for the creation of both the StreamReader and StreamWriter, you must set the

<route_types> tag to true.

In the following example, both the StreamWriter and StreamReader will be created as soon as the type ShapeType is discovered in either domain.

<topic_route> <route_types>true</route_types>

<input participant="1"> <creation_mode>IMMEDIATE</creation_mode> <registered_type_name>ShapeType</registered_type_name>

...

</input>

<output> <creation_mode>IMMEDIATE</creation_mode>

<registered_type_name>ShapeType</registered_type_name>

...

</output>

...

</topic_route>

In this next example, the StreamReader will be created only when the type ShapeType is discovered in the input domain; the StreamWriter will be created only when the type ShapeType is discovered in the output domain.

<topic_route> <route_types>false</route_types>

<input participant="1"> <creation_mode>IMMEDIATE</creation_mode> <registered_type_name>ShapeType</registered_type_name>

...

</input>

<output> <creation_mode>IMMEDIATE</creation_mode>

<registered_type_name>ShapeType</registered_type_name>

...

</output>

...

</topic_route>

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2-28

2.4.6.4Creation Modes—Controlling when StreamReaders and StreamWriters are Created

The way a route creates its StreamReader and StreamWriter and starts reading and writing data can be configured.

The <creation_mode> tag in a route’s <input> and <output> tags controls when the routing service StreamReader/StreamWriter is created. Table 2.14 lists the possible values for the

<creation_mode> tag.

Table 2.14 Creation Modes

<creation_mode> Values

Description

 

 

 

The route StreamReader/StreamWriter is created as soon as possible; that is, as

 

soon as the types are available. Note that if the type is defined in the

IMMEDIATE (default)

configuration file, the creation will occur when the routing service starts.

 

If the type is not defined in the configuration file, it has to be discovered; see

 

Discovering Types (Section 2.4.6.3).

 

 

 

The route StreamReader is not created until the associated connection

 

discovers a data Producer on the same stream.

 

For example, for a Connext domain, Routing Service will not create the route

 

DataReader until a DataWriter for the same topic is discovered on the same

ON_DOMAIN_MATCH

domain.

The routing service StreamWriter is not created until the associated connection

 

 

discovers a data Consumer on the same stream.

 

For example, for a Connext domain, Routing Service will not create the route

 

DataWriter until a DataReader for the same topic is discovered on the same

 

domain.

 

 

ON_ROUTE_MATCH

The routing service StreamReader/StreamWriter is not created until its

counterpart in the route is created.

 

 

 

ON_DOMAIN_AND_

Both conditions must be true.

ROUTE_MATCH

 

 

 

ON_DOMAIN_OR_

At least one of the conditions must be true.

ROUTE_MATCH

 

 

 

Route Destruction:

The same rules that are applied to create the route StreamWriter and StreamReader also apply to their destruction. When the condition that triggered the creation of that entity becomes false, the entity is destroyed. (Note that IMMEDIATE will never become false.)

For example, if the creation mode of a topic route's <input> tag is ON_DOMAIN_MATCH, when all the matching user DataWriters in the input domain are deleted, the topic route's DataReader is deleted.

Example 1

In this example, data is routed as soon as a user DataWriter is publishing it on the first domain.

<topic_route>

<input participant="1"> <creation_mode>ON_DOMAIN_MATCH</creation_mode>

...

</input>

<output> <creation_mode>

ON_ROUTE_MATCH

</creation_mode>

...

2-29

</output>

</topic_route>

Example 2

In this example, data is not routed until a user DataWriter is publishing and a user DataReader is already expecting it.

<topic_route>

<input participant="1"> <creation_mode>

ON_DOMAIN_AND_ROUTE_MATCH

</creation_mode>

...

</input>

<output> <creation_mode>

ON_DOMAIN_AND_ROUTE_MATCH

</creation_mode>

...

</output>

</topic_route>

Example 3

In this example, all the data is received by the topic route's DataReader, because it is created as soon as a user DataWriter is discovered on the first domain. However, the data is not resent until a user DataReader on the other domain subscribes to it.

<topic_route>

<input participant="1">

<creation_mode>

ON_DOMAIN_MATCH

</creation_mode>

...

</input>

<output> <creation_mode>

ON_DOMAIN_AND_ROUTE_MATCH

</creation_mode>

...

</output>

</topic_route>

2.4.6.5Data Transformation

A route can transform the incoming data using a transformation, an object created by a transformation plugin.

For example, the following transformation switches the coordinates of the input sample: x becomes y, and y becomes x.

<topic_route name="SquareSwitchCoord">

<input participant="1"> <topic_name>Square</topic_name>

2-30

<registered_type_name> ShapeType

</registered_type_name> </input>

<output> <topic_name>Square</topic_name> <registered_type_name>

ShapeType </registered_type_name>

</output>

<transformation plugin_name=”transformationLib::assign">

<property>

<value>

<element>

<name>X</name>

<value>Y</value>

</element>

<element>

<name>Y</name>

<value>X</value>

</element>

</value>

</property>

</transformation> </topic_route>

To include a transformation in a route:

1.Implement the transformation plugin API and generate a shared library. See Chapter 4: Transforming Data with Routing Service for more information.

2.Register that library in the configuration file by creating a <transformation_plugin> tag inside a <transformation_library> tag. (As noted in Table 2.1, <transformation_library> is a top-level tag.)

3.Instantiate a transformation by creating a <transformation> tag inside a <route> or a

<topic_route> tag.

Table 2.15 lists the tags allowed within a <transformation> tag.

For additional information about transformations see Chapter 4: Transforming Data with Routing Service.

2-31

Table 2.15 Transformation Tags

Tags within

Description

Number of

<transformation>

Tags Allowed

 

 

 

 

<input_type_name>

Type name of the data samples this transformation receives

0 or 1

 

 

 

<output_type_name>

Type name of the data samples this transformation creates

0 or 1

 

 

 

 

Sequence of name/value(string) pairs that can be used to

 

 

configure certain parameters of the transformation. For example:

 

 

<property>

 

 

<value>

 

<property>

<element>

0 or 1

<name>scaling_factor</name>

 

 

 

<value>2</value>

 

 

</element>

 

 

</value>

 

 

</property>

 

 

 

 

2.4.7Auto Routes

The tag <auto_route> defines a set of potential routes, with the same input and output type and same input and output stream name. A route can eventually be instantiated when a new stream is discovered with a type name and a stream name that match the filters in the auto route. When this happens, a route is created (but not necessarily started; see Section 2.4.6.4) with the configuration defined in the auto route tag.

For example:

<dds>

...

<routing_service>

...

<domain_route> <participant_1>

<domain_id>54</domain_id> </participant_1>

<connection_2 plugin_name=”my_adapter_library::jms> </connection_2>

...

<session>

...

<auto_route name="AutoRoute1">

...

<dds_input participant="1">

<allow_topic_name_filter>*

</allow_topic_name_filter> <allow_registered_type_name_filter>

ShapeType </allow_registered_type_name_filter>

...

</dds_input> <output>

<allow_stream_name_filter>A*

</allow_stream_name_filter>

<allow_registered_type_name_filter>

B*

</allow_registered_type_name_filter>

...

</output>

2-32

</auto_route>

...

</session>

...

</domain_route>

...

</routing_service>

...

</dds>

The above auto route will lead to the creation of a route every time any topic of type ShapeType is discovered on the Connext domain or a JMS queue/topic starting with A with a type starting with B is discovered on the output JMS connection.

For example, discovering the topic “Triangle” of “ShapeType” will trigger the creation of a topic route that routes triangles from the Connext domain to the JMS domain. Discovering a topic “Atopic” of type “Btype” on the JMS domain will trigger the creation of a topic route that routes “Atopic” from the Connext domain to the JMS domain.

Connext inputs and outputs within an auto route are defined using the XML tags <dds_input> and <dds_output>. Input and outputs from other data domains are defined using the tags

<input> and <output>.

An auto topic route is a special kind of route that defines a mapping between two Connext domains.

Please see the following tables for more information on allowable tags:

Table 2.16 on page 2-33 lists the tags allowed within an <auto_route> tag.

Table 2.17 on page 2-34 lists the tags allowed within an <auto_topic_route> tag.

Table 2.18 on page 2-34 lists the tags allowed within <input> and <output> tags nested within an <auto_route> tag.

Table 2.19 on page 2-36 lists the tags allowed within the <dds_input> and <dds_output> tags nested within an <auto_topic_route> or a <topic_route> tag.

Table 2.16 Auto Route Tags

Tags within

 

Number of

Description

Tags

<auto_route>

 

Allowed

 

 

 

 

 

 

 

 

 

Only applicable to Connext inputs.

 

<dds_input>

Defines the auto route’s input stream (topic). See Auto Routes

1 (required)

 

(Section 2.4.7)

 

 

 

 

 

Only applicable to Connext outputs.

 

<dds_output>

Defines the auto route’s output stream (topic). See Auto Routes

1 (required)

 

(Section 2.4.7).

 

 

 

 

 

Enables and configures remote monitoring for the auto route. See

 

<entity_monitoring>

Monitoring (Section 2.4.4) and Chapter 6: Monitoring Routing

0 or 1

 

Service from a Remote Location

 

 

 

 

 

Only applicable to non-Connext inputs.

 

<input>

Defines the auto route’s input stream. See Auto Routes (Section

1 (required)

 

2.4.7).

 

 

 

 

 

Only applicable to non-Connext outputs.

 

<output>

Defines the auto route’s output stream. See Auto Routes (Section

1 (required)

 

2.4.7).

 

 

 

 

2-33

Table 2.16 Auto Route Tags

 

Tags within

 

 

 

 

 

 

Number of

 

 

 

 

 

 

Description

 

Tags

 

<auto_route>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

When this tag is true, the data samples read from the input streams

 

 

 

 

 

 

 

are written into the output streams with the same timestamp that

 

 

 

<publish_with_

 

 

was associated with them when they were made available in the

 

 

 

 

 

input domain.

 

0 or 1

 

original_timestamp>

 

 

 

 

 

 

 

 

 

This option may not be applicable in some adapter’s

 

 

 

 

 

 

 

implementations where the concept of timestamp is not supported.

 

 

 

 

 

 

 

Default: false

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2.17 Auto-topic Route Tags

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tags within

 

 

 

 

 

 

Number of

 

 

 

 

 

 

Description

 

Tags

 

<auto_topic_route>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Enables and configures remote monitoring for the auto topic route.

 

 

 

<entity_monitoring>

 

 

 

See Monitoring (Section 2.4.4) and Chapter 6: Monitoring Routing

0 or 1

 

 

 

 

 

 

Service from a Remote Location

 

 

 

 

 

 

 

 

 

 

 

 

 

<input>

 

 

 

Defines the auto topic route’s input topic. See Auto Routes (Section

1 (required)

 

 

 

 

 

 

2.4.7).

 

 

 

 

 

<output>

 

 

 

Defines the auto topic route’s output topic. See Auto Routes

1 (required)

 

 

 

 

 

 

(Section 2.4.7).

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

 

 

 

 

 

When this flag is set to true, the NOT_ALIVE_DISPOSE samples

 

 

 

<propagate_dispose>

 

 

 

received by the topic routes’ DataReaders are not published with

0 or 1

 

 

 

 

 

 

the topic routes’ DataWriters.

 

 

 

 

 

 

 

 

Default: true

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

 

 

 

 

 

When this flag is set to true, the NOT_ALIVE_NO_WRITERS

 

 

 

<propagate_unregister>

 

samples received by the topic routes’ DataReaders are not

0 or 1

 

 

published with the topic routes’ DataWriters as unregister

 

 

 

 

 

 

samples.

 

 

 

 

 

 

 

 

 

Default: true

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

 

 

 

 

 

When this flag is set to true, if you have N topic routes for the same

 

 

 

<publish_with_

 

 

 

topic (in different routers or in the same one), each sample that was

0 or 1

 

original_info>

 

 

 

written from a DataWriter in the input domain will be routed N

 

 

 

 

 

 

times, but DataReaders on the output domain will only see one.

 

 

 

 

 

 

 

 

Default: false

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

<publish_with_

 

 

 

When this tag is set to true, the data samples are written with their

0 or 1

 

original_timestamp>

 

 

 

original source timestamp.

 

 

 

 

 

 

 

 

Default: false

 

 

 

 

 

 

 

 

 

 

Table 2.18 Input and Output Tags for the <auto_route> Tag

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tags within

 

Tags within

 

 

 

Number

 

 

 

Description

 

of Tags

 

<input>

 

 

 

<output>

 

 

 

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A registered type name filter.1

 

 

 

<allow_registered_type_name_filter>

 

You may use a comma-separated list to specify more

0 or 1

 

 

 

 

 

 

 

 

than one filter.

 

 

 

 

 

 

 

 

 

 

Default:* (allow all)

 

 

 

 

 

 

 

 

 

 

 

 

 

2-34

Table 2.18 Input and Output Tags for the <auto_route> Tag

Tags within

Tags within

 

 

 

 

 

 

Number

 

 

Description

 

 

of Tags

<input>

<output>

 

 

 

 

 

 

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A stream name filter.1

 

 

 

 

<allow_stream_name_filter>

You may use a comma-separated list to specify more

0 or 1

 

 

than one filter.

 

 

 

 

 

 

 

 

Default:* (allow all)

 

 

 

 

 

 

 

 

 

 

 

The routes are created with this configuration. See

 

<creation_mode>

 

Creation Modes—Controlling when

StreamReaders

0 or 1

 

 

and StreamWriters are Created (Section 2.4.6.4).

 

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

 

The contents of this tag are specified in the same

 

<datareader_qos>

<datawriter_qos>

manner as for a Connext QoS profile file—see Chapter

0 or 1

15 in the RTI Core Libraries and Utilities User’s Manual.2

 

 

If the tag is not defined, Routing Service will use the

 

 

 

Connext defaults.

 

 

 

 

 

 

 

 

 

 

 

A registered type name filter1 that should be denied

 

 

 

(excluded).

This

is

applied

after

the

 

<deny_registered_type_name_filter>

<allow_registered_type_name_filter>.

 

 

0 or 1

You may use a comma-separated list to specify more

 

 

than one filter.

 

 

 

 

 

 

 

 

Default: Not applied

 

 

 

 

 

 

 

 

 

 

 

A stream name filter1 that should be denied

 

 

 

(excluded).

This

is

applied

after

the

 

<deny_stream_name_filter>

<allow_stream_name_filter>.

 

 

 

0 or 1

You may use a comma-separated list to specify more

 

 

than one filter.

 

 

 

 

 

 

 

 

Default:* (allow all)

 

 

 

 

 

 

 

 

 

 

 

A topic name filter1 that should be denied (excluded).

 

 

 

This is applied after the <allow_topic_name_filter>.

 

<deny_topic_name_filter>

You may use a comma-separated list to specify more

0 or 1

 

 

than one filter.

 

 

 

 

 

 

 

 

Default: Not applied

 

 

 

 

 

 

 

 

 

 

 

 

The topic routes are created with this configuration.

 

 

 

 

Sequence of name/value(string) pairs that can be used

 

 

 

to configure certain parameters of the StreamReaders/

 

 

 

StreamWriters associated with the routes created from

 

 

 

the auto route. For example:

 

 

 

 

 

 

<property>

 

 

 

 

 

 

<property>

 

<value>

 

 

 

 

 

0 or 1

 

 

<element>

 

 

 

 

 

 

 

 

<name> com.rti.socket.port</name>

 

 

 

 

 

<value>16556</value>

 

 

 

 

 

 

</element>

 

 

 

 

 

 

 

 

</value>

 

 

 

 

 

 

 

 

</property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.As defined by the POSIX fnmatch API (1003.2-1992 section B.6).

2.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2-35

Table 2.19 Connext Input and Output Tags for the <auto_topic_route> and <auto_route> Tags

Tags within

Tags within

 

 

 

 

 

Number

<auto_topic_route><input>

<auto_topic_route><output>

 

 

 

 

 

 

Description

 

of Tags

and

and

 

 

 

 

 

 

 

Allowed

<auto_route><dds_input>

<auto_route><dds_output>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A registered type name filter.1

 

<allow_registered_type_name_filter>

You may use a comma-separated list

0 or 1

 

 

to specify more than one filter.

 

 

 

Default:* (allow all)

 

 

 

 

 

 

 

 

 

 

 

A topic name filter.1

 

 

 

<allow_topic_name_filter>

 

You may use a comma-separated list

0 or 1

 

 

to specify more than one filter.

 

 

 

Default:* (allow all)

 

 

 

 

 

 

 

 

 

The topic routes are created with a

 

 

 

SQL content filter topic with this

 

 

 

expression.

 

 

 

 

 

 

<auto_topic_route>

 

 

 

 

 

...

 

 

 

 

 

 

 

<input>

 

 

 

 

 

 

...

 

 

 

 

<content_filter>

N/A

<content_filter>

 

 

0 or 1

 

<expression>

 

 

 

 

 

 

 

 

 

 

 

x > 100

 

 

 

 

 

 

 

</expression>

 

 

 

 

 

</content_filter>

 

 

 

 

 

 

...

 

 

 

 

 

 

</input>

 

 

 

 

 

 

...

 

 

 

 

 

 

 

</auto_topic_route>

 

 

 

 

 

 

 

 

 

The topic routes are created with

 

 

 

this

configuration.

See

Creation

 

<creation_mode>

 

Modes—Controlling

 

when

0 or 1

 

 

StreamReaders

and

StreamWriters

 

 

 

are Created (Section 2.4.6.4).

 

 

 

 

 

 

 

The topic routes are created with

 

 

 

this configuration.

 

 

 

 

 

The contents of this tag are specified

 

 

 

in the same manner as for a Connext

 

<datareader_qos>

<datawriter_qos>

QoS profile file—see Chapter 15 in

0 or 1

 

 

the RTI Core Libraries and Utilities

 

 

 

User’s Manual.2

 

 

 

 

 

 

If the tag is not defined, Routing

 

 

 

Service will use the Connext defaults.

 

 

 

 

 

 

 

A registered type name filter1 that

 

 

 

should be denied (excluded).

 

 

 

This

is

applied

after

 

<deny_registered_type_name_filter>

<allow_registered_type_name_filt

0 or 1

er>.

 

 

 

 

 

 

 

 

 

 

 

 

 

You may use a comma-separated list

 

 

 

to specify more than one filter.

 

 

 

Default: Not applied

 

 

 

 

 

 

 

 

 

 

 

2-36

Table 2.19 Connext Input and Output Tags for the <auto_topic_route> and <auto_route> Tags

Tags within

Tags within

 

 

Number

<auto_topic_route><input>

<auto_topic_route><output>

 

 

Description

 

of Tags

and

and

 

 

 

Allowed

<auto_route><dds_input>

<auto_route><dds_output>

 

 

 

 

 

 

 

 

 

 

 

A topic name filter1 that should be

 

 

 

denied (excluded).

 

 

 

 

This is applied after

the

 

<deny_topic_name_filter>

 

<allow_topic_name_filter>.

 

0 or 1

 

 

You may use a comma-separated list

 

 

 

to specify more than one filter.

 

 

 

 

Default: Not applied

 

 

 

 

 

 

 

1.As defined by the POSIX fnmatch API (1003.2-1992 section B.6).

2.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

2.4.8Adapters

Adapters are pluggable components that allow Routing Service to consume and produce data for different data domains (e.g., Connext, JMS, Socket, etc.). By default, Routing Service is distributed with a built-in Connext adapter. Any other adapters must be registered within the

<adapter_library> tag.

To support new data domains:

1.Implement the adapter plugin API in Java or C. See Chapter 8: Extending Routing Ser- vice with Adapters for more information.

2.Register the plugin in the configuration file by creating an <adapter_plugin> tag or a

<java_adapter_plugin> inside an <adapter_library> tag. (As noted in Table 2.1, <adapter_library> is a top-level tag.)

3.Instantiate an adapter connection by creating a <connection> tag inside a <domain_route> tag that refers to the adapter plugin.

For additional information about adapters see Chapter 8: Extending Routing Service with Adapters.

2.5Enabling and Disabling Routing Service Entities

The Routing Service entities associated with the tags <routing_service>, <domain_route>,

<route>, <topic_route>, <auto_route>, and <auto_topic_route> can be created enabled or disabled using the attribute enabled.

By default, the value of the enabled attribute is true.

For example:

<dds>

<routing_service name="TopicBridgeExample"

group_name="rti.router.default" enabled="true">

<domain_route name="DomainRoute" enabled ="false">
<participant_1>

<domain_id>0</domain_id> </participant_1>

<participant_2> <domain_id>1</domain_id>

2-37

</participant_2>

<session name="Session">

<topic_route name="SquaresToCircles" enabled="false"> <input participant="1">

<registered_type_name>ShapeType </registered_type_name> <topic_name>Square</topic_name>

</input>

<output> <registered_type_name>ShapeType </registered_type_name> <topic_name>Circle</topic_name>

</output> </topic_route> </session>

</domain_route> </routing_service>

</dds>

When an entity is created disabled, it can be enabled remotely using the commands enable (Section 5.2.5) and disable (Section 5.2.4). A routing_service can be created disabled by setting the attribute enabled to false or by using the -noAutoEnable command-line option. The command-line parameter takes precedence over the XML attribute value.

2.6Enabling RTI Distributed Logger in Routing Service

Routing Service provides integrated support for RTI Distributed Logger.

When you enable Distributed Logger, Routing Service will publish its log messages to Connext. Then you can use RTI Monitor1 to visualize the log message data. Since the data is provided in a Connext topic, you can also use rtiddsspy or even write your own visualization tool.

To enable Distributed Logger, modify the Routing Service XML configuration file. In the <administration> section, add the <distributed_logger> tag as shown in the example below.

<routing_service name="default">

<administration>

...

<distributed_logger>

<enabled>true</enabled>

</distributed_logger>

</administration>

...

</routing_service>

There are more configuration tags that you can use to control Distributed Logger’s behavior. For example, you can specify a filter so that only certain types of log messages are published. For details, see the RTI Distributed Logger Getting Started Guide.

1. RTI Monitor is a separate CUI application that can run on the same host as your application or on a different host.

2-38

2.7Support for Extensible Types

Routing Service includes partial support for the "Extensible and Dynamic Topic Types for DDS" specification from the Object Management Group (OMG)1. This section assumes that you are familiar with Extensible Types and you have read the Core Libraries and Utilities Getting Started Guide Addendum for Extensible Types.

Topic Routes can subscribe to and publish topics associated with final and extensible types.

You can select the type version associated with a topic route by providing the type description in the XML configuration file. The XML description supports structure inheritance. If you have the Connext Core Libraries and Utilities, you can learn more about

structure inheritance in the Core Libraries and Utilities Getting Started Guide Addendum for Extensible Types2.

The TypeConsistencyEnforcementQosPolicy can be specified on a per-topic-route basis, in the same way as other QoS policies.

Within a domain_route, a topic cannot be associated with more than one type version. This prevents the same domain route from having two topic routes with different versions of a type for the same Topic. To achieve this behavior, create two different domain routes, each associating the topic with a different type version.

The type declared in a topic route input is the version that is passed to the output (or to a transformation). The topic route can subscribe to different-but-compatible types, but those samples are translated to the topic route’s input type.

For example:

struct A { long x;

};

struct B {

long x; long y;

};

Samples published by

Samples forwarded by a Routing Service

Samples received

topic route for type A

two writers of types A

 

 

by a B reader

and B, respectively

Input

Output

 

 

 

 

 

 

 

A [x=1]

A [x=1]

A [x = 1]

B [x=1, y=0]

 

A [x=10]

A [x=10]

B [x=10, y=0]

B [x=10, y=11]

 

 

 

 

Note that the second sample loses the extended field when it is forwarded by Routing Service. A topic route using the extended type would avoid that truncation:

 

Samples published by

Samples forwarded by a Routing Service topic

Samples received

 

 

 

route for type B

 

two writers of types A

 

 

 

 

 

 

 

by a B reader

 

and B, respectively

 

Input

 

Output

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A [x=1]

B [x=1, y=0]

 

B [x = 1, y=0]

B [x=1, y=0]

 

 

 

 

 

 

 

 

B [x=10, y=11]

B [x=10, y=11]

 

B [x=10, y = 11]

B [x=10, y=11]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.http://www.omg.org/spec/DDS-XTypes/

2.<Connext installation directory>/ndds.<version>/doc/pdf/

RTI_CoreLibrariesAndUtilities_GettingStarted_ExtensibleTypesAddendum.pdf)

2-39

2.7.1Example

The following XML configuration file showcases the features mentioned in the previous section.

<?xml version="1.0"?>

<dds xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xsi:noNamespaceSchemaLocation= "resource/schema/ rti_routing_service.xsd">

<types>

<!-- Base type -->

<struct name="MyBaseType" extensibility="extensible"> <member name="x" type="long"/>

<member name="y" type="long"/>

</struct>

<!-- Extended type (structure inheritance) -->

<struct name="MyDerivedType" baseType="MyBaseType" extensibility="extensible">

<member name="z" type="long"/>
</struct>

</types>

<routing_service name="ExtensibleTypesTest">

<domain_route name="test_dr"> <participant_1>

<domain_id>0</domain_id>

<!-- Use of type signature --> <registered_type name="MyDerivedType"

type_name="MyDerivedType"/>

</participant_1>

<participant_2> <domain_id>1</domain_id> <!-- Use of type signature -->

<registered_type name="MyDerivedType" type_name="MyDerivedType"/>

</participant_2>

<session name="test_s" enabled="true"> <topic_route name="derived_tr">

<input participant="1"> <topic_name>MyTopic</topic_name>

<!-- Refer to the registered type name --> <registered_type_name>

MyDerivedType </registered_type_name>

2-40

<datareader_qos>

<!-- Define DataReader TypeConsistencyEnforcementQos -->

<type_consistency> <kind>

ALLOW_TYPE_COERCION </kind>

</type_consistency> </datareader_qos>

</input>

<output>

<topic_name>MyTopic</topic_name>

<!-- Refer to the registered type name -->

<registered_type_name>

MyDerivedType
</registered_type_name>

</output>
</topic_route>

</session>
</domain_route>

</routing_service>
</dds>

2-41

Chapter 3 Running Routing Service

3.1Starting Routing Service

Routing Service runs as a separate application. The script to run the executable is located in

<Routing Service installation directory>/scripts.

Routing Service supports loading Java adapters. If your configuration is set up to load a Java adapter, follow these steps:

1.On Windows Systems: To use a Java adapter, you must have the Visual Studio 2005

service pack 1 redistributable libraries. You can obtain this package from Microsoft or RTI (see the RTI Core Libraries and Utilities Release Notes1 for details).

2.Make sure Java 1.5 or higher is available.

3.Make sure you add the directory of the Java Virtual Machine dynamic library to your environment variable: LD_LIBRARY_PATH (on UNIX-based systems) or Path (on Windows systems). For example:

setenv LD_LIBRARY_PATH \

${LD_LIBRARY_PATH}:/local/java/jdk1.5.0_07/jre/lib/i386/client

To start Routing Service, enter:

cd <Routing Service installation directory> scripts/rtiroutingservice [options]

Example:

cd <Routing Service installation directory>

scripts/rtiroutingservice -cfgFile example/shapes/topic_bridge.xml -cfgName example

Table 3.1 describes the command-line options.

3.2Stopping Routing Service

To stop Routing Service, press Ctrl-c. Routing Service will perform a clean shutdown.

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_ReleaseNotes.pdf.

3-1

Table 3.1 RTI Routing Service Command-line Options

Option

Description

 

 

 

 

 

Assigns a name to the execution of the Routing Service.

 

Remote commands and status information will refer to the routing

 

service using this name. See the Routing Service User’s Manual for more

-appName <name>

information.

In addition, the name of DomainParticipants created by Routing

 

 

Service will be based on this name.

 

Default: The name given with -cfgName, if present, otherwise it is

 

“RTI_Routing_Service”.

 

 

-cfgFile <name>

Specifies a configuration file to be loaded.

See How to Load the XML Configuration (Section 2.2).

 

 

 

 

Specifies a configuration name. Routing Service will look for a matching

-cfgName <name>

<routing_service> tag in the configuration file.

This parameter is required unless you use

 

 

-remoteAdministrationDomainId and -noAutoEnable.

 

 

 

Sets the base domain ID.

 

This value is added to the domain IDs in the configuration file. For

-domainIdBase <ID>

example, if you set -domainIdBase to 50 and use domain IDs 0 and 1

in the configuration file, then the Routing Service will use domains 50

 

 

and 51.

 

Default: 0

 

 

-help

Displays help information.

 

 

 

Appends the host name and process ID to the service name provided

-identifyExecution

with the -appName option. This helps ensure unique names for remote

administration and monitoring.

 

 

For example: MyRoutingService_myhost_20024

 

 

 

Specifies the license file (path and filename). Only applicable to

-licenseFile <file>

licensed versions of Routing Service.

If not specified, Routing Service looks for the license as described in

 

 

Installing the License File (Section 2.3) in the Getting Started Guide.

 

 

-maxObjectsPerThread <int>

Parameter for the DomainParticipantFactory.

 

 

 

Starts Routing Service in a disabled state.

 

Use this option if you plan to enable Routing Service remotely, as

-noAutoEnable

described in the User’s Manual.

 

This option overwrites the value of the enable attribute in the

 

<routing_service> tag.

 

 

 

Enables remote administration and sets the domain ID for remote

 

communication.

 

When remote administration is enabled, Routing Service will create a

 

DomainParticipant, Publisher, Subscriber, DataWriter, and DataReader in

-remoteAdministrationDomainId

the designated domain. The QoS values for these entities are described

in the Routing Service User’s Manual.

<ID>

This option overwrites the value of the tag <domain_id> within a

 

 

<administration> tag. (See the Routing Service User’s Manual for

 

information on configuring remote access).

 

Default: Remote administration is not enabled unless it is enabled from

 

the XML file.

 

 

3-2

Table 3.1 RTI Routing Service Command-line Options

Option

 

Description

 

 

 

 

 

Enables remote monitoring and sets the domain ID for status

 

publication.

 

When remote monitoring is enabled, Routing Service will create one

 

DomainParticipant, one Publisher, five DataWriters for data

 

publication (one for each kind of entity), and five DataWriters for

-remoteMonitoringDomainId <ID>

status publication (one for each kind of entity). The QoS values for

these entities are described in the Routing Service User’s Manual.

 

This option overwrites the value of the tag <domain_id> within a

 

<monitoring> tag. (See the Routing Service User’s Manual for

 

information on configuring remote monitoring).

 

Default: Remote monitoring is not enabled unless it is enabled from

 

the XML file.

 

 

-stopAfter <sec>

Stops the service after the specified number of seconds.

 

 

 

Enables compatibility with RTI Data Distribution Service 4.2e.

 

This option should be used when compatibility with 4.2e is required

-use42eAlignment

and the topic data types contain double, long long, unsigned long

 

long, or long double members.

 

Default: Disabled

 

 

 

Controls what type of messages are logged:

 

0

- Silent

 

1

- Exceptions (Connext and Routing Service) (default)

 

2

- Warnings(Routing Service)

-verbosity <n>

3 - Information (Routing Service)

 

4

- Warnings (Connext and Routing Service)

 

5

- Tracing (Routing Service)

 

6

- Tracing (Connext and Routing Service)

 

Each verbosity level, n, includes all the verbosity levels smaller than n.

 

 

-version

Prints the Routing Service version number.

 

 

 

3-3

Chapter 4 Transforming Data with Routing Service

As described in Data Transformation (Section 2.4.6.5), a route can transform the incoming data using a transformation, which is an object created by a transformation plugin.

Transformation plugins implement the transformation API and must be provided as shared libraries that Routing Service will load dynamically.

Currently, the transformation plugin API is only supported in C.

This chapter describes:

Transformation Usage and Configuration (Section 4.1)

Transformations Distributed with Routing Service (Section 4.2)

Creating New Transformations (Section 4.3)

4.1Transformation Usage and Configuration

In the XML configuration file, transformation plugins must be defined within a transformation library.

For example:

<dds>

<transformation_library name="MyTransfLib">

<transformation_plugin name="MyTransfPlugin">

<dll>mytransformation</dll>

<create_function>MyTransfPlugin_create</create_function>

</transformation_plugin>

...

</transformation_library>

...

<routing_service>

...

</routing_service>

...

</dds>

Table 4.1 on page 4-2 lists the tags allowed within <transformation_plugin>.

Table 4.2 on page 4-3 lists the tags allowed within a <transformation> tag.

4-1

Once a transformation plugin is registered, a route can use it to create a data transformation. For example, the following route uses a transformation to switch the coordinates of the input sample: x becomes y, and y becomes x.

<topic_route name="SquareSwitchCoord"> <input participant="1">

<topic_name>Square</topic_name> <registered_type_name>ShapeType</registered_type_name>

</input>

<output> <topic_name>Square</topic_name>

<registered_type_name>ShapeType</registered_type_name> </output>

<transformation plugin_name=”MyTransfLib::MyTransPlugin">

<property>

<value>

<element>

<name>X</name>

<value>Y</value>

</element>

<element>

<name>Y</name>

<value>X</value>

</element>

</value>

</property>

</transformation> </topic_route>

Table 4.1 Transformation Plugin Tags

Tags within

 

Number

<transformation_

Description

of Tags

plugin>

 

Allowed

 

 

 

 

Shared library containing the implementation of the transformation plugin.

 

 

The <dll> tag may specify the exact name of the file (for example, lib/lib-

 

 

mytransformation.so) or a general name (no file extension) which will be

 

 

completed as follows:

 

 

<dll> value: dir/mytransformation

 

<dll>

Final Path (UNIX-based systems): dir/libmytransformation.so

1

Final Path (Windows systems): dir/mytransformation.dll

(required)

 

 

If the library specified with the <dll> tag cannot be opened (because the

 

 

library path is not in the Path environment variable on Windows or the

 

 

LD_LIBRARY_PATH environment variables on UNIX-based systems),

 

 

Routing Service will look for the library in <Routing Service installation

 

 

directory>/bin/<architecture>.

 

 

 

 

<create_function>

This tag will contain the name of the function used to create the transfor-

1

mation plugin (see Section 4.3.1).

(required)

 

The function must be implemented in the shared library.

 

 

 

 

 

4-2

Table 4.2 Transformation Tags

Tags within

 

Number

Description

of Tags

<transformation>

 

Allowed

 

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure the

 

 

parameters of the transformation. For example:

 

 

<property>

 

 

<value>

 

 

<element>

 

 

<name>X</name>

 

 

<value>Y</value>

 

 

</element>

 

<property>

<element>

0 or 1

 

<name>Y</name>

 

 

<value>X</value>

 

 

</element>

 

 

</value>

 

 

</property>

 

 

In this example, the properties are used to define the field assignments. The

 

 

semantics associated with the transformation property value depends on the

 

 

plugin implementation.

 

 

 

 

4.2Transformations Distributed with Routing Service

Routing Service provides a transformation that is able to map fields of the input type to fields of the output type using the property tag inside the transformation to provide this mapping. For example:

<dds>

...

<transformation_library name="TransformationLib"> <transformation_plugin name="Assignment"> <dll>rtirsassigntransf</dll>

<create_function> RTI_RoutingServiceAssignTransformationPlugin_create

</create_function> </transformation_plugin>

...

</transformation_library>

...

<routing_service name=”MyService”> <domain_route name=”MyDomainRoute”> <session name=”MySession”>

<route name=”MyRoute”>

...

<transformation plugin_name= "TransformationLib::Assignment">

<property>

<value>

<element>

<name>X</name>

<value>Y</value>

</element>

<element>

<name>Y</name>

4-3

<value>X</value>

</element>

</value>

</property>

</transformation>

</route>

...

</session>

...

</domain_route>

...

</routing_service>

...

</dds>

This transformation plugin is implemented in the shared library, <Routing Service installation directory>/bin/<architecture>/librtirsassigntransf.so (or rtirsassigntransf.dll for Windows systems).

Important:

The assign transformation only supports the assignment of primitive fields (including strings) that are not part of arrays or sequences. For example:

<transformation plugin_name= "TransformationLib::Assignment">

<property>

<value>

<element>

<name>position.x</name>

<value>position.y</value>

</element>

<element> <!-- not supported -->
<name>x[0]</name> <value>y[0]</value>

</element>

<element> <!-- supported -->
<name>position</name> <value>position</value>

</element>

</value>

</property>

</transformation>

4.3Creating New Transformations

Routing Service provides a transformation SDK in C to support the creation of custom transformation plugins.

The SDK contains two main components:

API header file: <Routing Service installation directory>/include/routingservice/ routingservice_transformation.h.

The transformation plugin will include this header.

Infrastructure library: <Routing Service installation directory>/lib/<architecture>/ librtirsinfrastructure.so (for UNIX-based systems) and <Routing Service installation directory>/lib/<architecture>/rtirsinfrastructure.dll (for Windows systems).

4-4

The transformation plugin will link with this library.

Transformation plugins working with TypeCode and DynamicData must also link with the Connext libraries.

Important: Because RTI only distributes the release version of Routing Service, your transformation should be linked against the release version of the Connext shared libraries when needed.

4.3.1Transformation Plugin API

Every transformation plugin will implement a plugin constructor (entry point to the shared library) that will be used by Routing Service to create a plugin instance.

typedef struct RTI_RoutingServiceTransformationPlugin *

(*RTI_RoutingServiceTransformationPlugin_create)(

RTI_RoutingServiceEnvironment * env);

The structure RTI_RoutingServiceTransformationPlugin will contain the plugin implementation as a set of function pointers.

struct RTI_RoutingServiceTransformationPlugin { RTI_RoutingServiceTransformationPlugin_DeleteFcn

transformation_plugin_delete; RTI_RoutingServiceTransformationPlugin_CreateTransformationFcn

transformation_plugin_create_transformation; RTI_RoutingServiceTransformationPlugin_DeleteTransformationFcn

transformation_plugin_delete_transformation; RTI_RoutingServiceTransformation_TransformFcn

transformation_transform; RTI_RoutingServiceTransformation_ReturnLoanFcn

transformation_return_loan; RTI_RoutingServiceConfigurableEntity_UpdateFcn

transformation_update; void * user_object;

};

The rest of this section introduces the different transformation functions. For detailed information about the API, please see the online (HTML) Routing Service documentation.

delete

Deletes the transformation plugin instance.

typedef void (*RTI_RoutingServiceTransformationPlugin_delete)(

struct RTI_RoutingServiceTransformationPlugin * plugin,

RTI_RoutingServiceEnvironment * env);

create_transformation

Creates a new transformation. The function is called when the route containing the transformation is ready to forward data.

typedef RTI_RoutingServiceTransformation (*RTI_RoutingServiceTransformationPlugin_create_transformation)(

struct RTI_RoutingServiceTransformationPlugin * plugin,

const struct RTI_RoutingServiceTypeInfo * input_type_info,

const struct RTI_RoutingServiceTypeInfo * output_type_info,
const struct RTI_RoutingServiceProperties * properties,
RTI_RoutingServiceEnvironment * env);

The behavior of the transformation can be configured using the properties parameter.

4-5

delete_transformation

Deletes a transformation. The function is called when the route containing the transformation is disabled.

typedef void (*RTI_RoutingServiceTransformationPlugin_delete_transformation)(

struct RTI_RoutingServiceTransformationPlugin * plugin, RTI_RoutingServiceTransformation transformation, RTI_RoutingServiceEnvironment * env);

The transformation parameter corresponds to the value returned by the function create_transformation().

transform

This function is called in a route to transform a sequence of input data samples into a sequence of output data samples.

typedef void (*RTI_RoutingServiceTransformation_transform)
( RTI_RoutingServiceTransformation transformation,
RTI_RoutingServiceSample ** out_sample_lst,
RTI_RoutingServiceSampleInfo ** out_info_lst,
unsigned int * out_count,
RTI_RoutingServiceSample * in_sample_lst,
RTI_RoutingServiceSampleInfo * in_info_lst,
unsigned int in_count,
RTI_RoutingServiceEnvironment * env);

When the routing service is done using the transformation by calling the return_loan()

output samples, it will ‘return the loan' to the operation.

The transformation parameter corresponds to the value returned by the function create_transformation().

return_loan

Indicates to the transformation that the routing service is done accessing the sequence of data samples obtained by an earlier invocation of transform().

typedef void (*RTI_RoutingServiceTransformation_return_loan)
( RTI_RoutingServiceTransformation transformation,
RTI_RoutingServiceSample * sample_lst,
RTI_RoutingServiceSampleInfo * info_lst,
unsigned int count,
RTI_RoutingServiceEnvironment * env);

The transformation parameter corresponds to the value returned by the function create_transformation().

update

This function is called when the configuration of a transformation changes as a result of a remote update command.

typedef void (*RTI_RoutingServiceTransformation_UpdateFcn)(

RTI_RoutingServiceTransformation transformation,

const struct RTI_RoutingServiceProperties * properties,

RTI_RoutingServiceEnvironment * env);

4-6

Chapter 5 Administering Routing Service from a Remote Location

Routing Service can be controlled remotely by sending commands through a special Connext Topic. Any Connext application can be implemented to send these commands and receive the corresponding responses. A shell application that sends/receives these commands is provided with Routing Service.

The script for the shell application is in <Routing Service installation directory>/scripts/rtirssh.

Entering rtirssh -help will show you the command-line options:

RTI Routing Service Shell

Usage: rtirssh [options]...

Options:

-domainId <integer>

Domain id for the remote configuration

-timeout

<seconds>

Max time to wait a remote response

-cmdFile

<file> Run commands in this file

-help

Displays this information

5.1Enabling Remote Administration

By default, remote administration is disabled in Routing Service for security reasons.

To enable remote administration you can use the <administration> tag (see Section 2.4.3) or the -remoteAdministrationDomainId <ID> command-line parameter, which enables remote administration and sets the domain ID for remote communication. For more information about the command-line options, see Starting Routing Service (Section 3.1).

When remote administration is enabled, Routing Service will create a DomainParticipant, Pub- lisher, Subscriber, DataWriter, and DataReader in the designated domain. (The QoS values for these entities are described in Section 2.4.3.)

5-1

5.2Remote Commands

This section describes the remote commands using the shell interface; Section 5.3 explains how to use remote administration from a Connext application.

Remote commands:

add_peer <target_routing_service> <domain_route_name> p1|p2 <peer_list>

create <target_routing_service> domain_route|session|topic_route|auto_route [<parent_entity_name>] <xml_url> [remote|local]

delete <target_routing_service> [<entity_name>]

disable <target_routing_service> [<entity_name>]

enable <target_routing_service> [<entity_name>]

get <target_routing_service>

load <target_routing_service> <cfg_name><xml_url> [remote|local]

pause <target_routing_service> [<entity_name>]

resume <target_routing_service> [<entity_name>]

save <target_routing_service>

update <target_routing_service> [<entity_name>] [<xml_url>|<assignment_expr>] [remote|local]

Parameters:

<assignment_expr> can be used instead of <xml_url> to modify single values in an entity configuration.

The assignment expression has the form: <fully qualified value name> = <value> For example:

update ShapeRouter DomainRoute1::Session1::SquareToCircles topic_route.input.datareader_qos.deadline.period.sec = 3

update ShapeRouter DomainRoute1::Session1::SquareToCircles topic_route.input.content_filter.expression = “x < 30”

<domain_route_name> is the fully qualified name of a domain route entity

<entity_name> is a fully qualified name. For example, consider the following XML config- uration:

<routing_service name="ShapeRoutingService">

...

<domain_route name="DomainRoute1">

...

<session name="Session">

<topic_route name="SquaresToCircles">

...

The above XML configuration would allow you to use commands such as:

enable ShapeRoutingService DomainRoute1::Session::SquaresToCircles

enable ShapeRoutingService DomainRoute1

Note that the fully qualified name does not include the name of the routing service.

<peer_list> is a comma-separate list of peers

5-2

<target_routing_service> can be:

The application name of a routing service, such as “MyRoutingService1”, as specified at start-up with the command-line option -appName

A regular expression1 for a routing service name, such as “MyRoutingService*”

<xml_url> can be:

A file URL, such as file:///home/user/myconfig.xml

A string URL, such as:

str://"<topic_route><input><datareader_qos>...

</datareader_qos></input></topic_route>"

If you omit the URL schema name, Routing Service will assume a file name; for example,

/home/user/myconfig.xml is equivalent to file:///home/user/myconfig.xml.

In either case, the XML code can represent either a whole DTD-valid configuration file (such as the one you specify through the command-line when you start Routing Service) or a snippet of XML that only refers to a specific entity (this is further explained in Section 5.2.12).

The [remote|local] parameter is used with file URLs to indicate if the file is local to the shell (local) or local to the routing service (remote). If the file is local to the shell (local), the shell application will read it and will send it as a string URL. If the file is local to the routing service (remote), the shell will send it as a file URL that will be read by the rout- ing service. The default value is remote.

If a relative path is specified, the path will be relative to the working directory in which the routing service (if remote is specified) or shell (if local is specified) is running.

5.2.1add_peer

add_peer <target_routing_service> <domain_route_name> p1|p2 <peer_list>

The add_peer command passes the peer_list to the underlying DomainParticipant's add_peer() function. It is only valid for DomainParticipants in a domain route.

<domain_route_name> is like <entity_name>, but must be a domain route entity.

p1|p2 specifies if the DomainParticipant associated with <participant_1> or <participant_2> configuration is selected.

<peer_list> is a comma-separated list of peers.

5.2.2create

create <target_routing_service> domain_route|session|topic_route|auto_route [<parent_entity_name>] <xml_url> [remote|local]

The create command is similar to update (Section 5.2.12), but the configuration is applied to a newly created entity instead of an existing one.

The second parameter (domain_route|session|topic_route|auto_route) is the kind of entity to be created. If the kind is a domain_route, there will be no parent. For the other kinds (session, topic_route, or auto_route), a <parent_entity_name> must be specified.

<xml_url> and [remote|local] are the same as used in update (Section 5.2.12), except that only XML snippets matching the entity kind are allowed. A full file (starting with <dds>...) is not valid.

1. As defined by the POSIX fnmatch API (1003.2-1992 section B.6)

5-3

For example (this would be entered as a single command, with no line-breaks):

create example topic_route DomainRoute::Session str://"<topic_route name="TrianglesToTriangles">

<input participant="1"><registered_type_name>ShapeType </registered_type_name><topic_name>Triangle</topic_name></input> <output><registered_type_name>ShapeType</registered_type_name> <topic_name>Triangle</topic_name></output></topic_route>"

5.2.3delete

delete <target_routing_service> [<entity_name>]

You can invoke the delete command on domain routes, routes and auto routes. It acts like the disable (Section 5.2.4) command, but also purges the configuration data for the target entity.

For example:

delete example DomainRoute::Session::CirclesToCircles

A deleted entity cannot be re-enabled, but a new one can be created.

5.2.4disable

disable <target_routing_service> [<entity_name>]

The disable command disables a routing service entity by destroying its sub-entities and corre- sponding Connext objects:

Routing service—When a routing service is disabled, all of its domain routes are destroyed. You do not need to specify the entity_name to disable a routing service.

Domain route—When a domain route is disabled, all its routes, topic routes, auto routes, and auto topic routes are destroyed, as well as both Connections (DomainParticipants for Connext). All the session threads are stopped and their corresponding adapter sessions (Publisher and Subscriber for Connext) are also deleted.

Route, topic route, auto route and auto topic route—When a route, topic route, auto route, or auto topic route is disabled, its StreamReaders and StreamWriters are destroyed, so data will no longer be routed.

5.2.5enable

enable <target_routing_service> [<entity_name>]

The enable command enables an entity that has been disabled or marked as ‘enabled=false’ in the configuration file.

This command can be used to enable the following entities:

Routing service—When a routing service is enabled, it uses the currently loaded configu- ration and starts. You don't need to specify the entity_name to enable a routing service.

Domain route—When a domain route is enabled, it creates the participants, routes, topic routes, auto routes, and auto topic routes that it contains. The routes, topic routes, auto routes, and auto topic routes will be created enabled or disabled depending on their cur- rent configuration. Enabling a domain route is required to start routing data from the input domain to the output domain.

Route, topic route, auto route, and auto topic route—Enabling a route, topic route, auto route or auto topic route is a necessary condition to start routing data between input and output streams. However, data routing will not start until the StreamWriter and Stream- Reader associated with a route are created (see Section 2.4.6.4 for additional informa- tion).

5-4

5.2.6get

get <target_routing_service>

The get command retrieves the current configuration.

The retrieved configuration, provided in an XML string format, is functionally equivalent to the loaded XML file, plus any updates (either from an update command or other remote commands that change the configuration, such as add_peer). However, the retrieved configuration may not be textually equivalent. For example, the retrieved configuration may explicitly contain default values that were not in the initial XML.

5.2.7load

load <target_routing_service> <cfg_name> <xml_url> [remote|local]

The load command loads specific XML configuration code. The target_routing_service must be disabled. For more information, see How to Load the XML Configuration (Section 2.2).

The XML code received must represent a valid routing service configuration file. The name of the <routing_service> tag to load is identified with <cfg_name>.

5.2.8pause

pause <target_routing_service> <entity_name>

When the pause command is called in a route, the session thread containing this route will stop reading data from the route’s StreamReader.

For routing service, domain routes, auto routes, and auto topic routes, the execution of this com- mand will pause the contained topic routes and routes.

5.2.9resume

resume <target_routing_service> <entity_name>

When the resume command is called in a route, the session thread containing this route will con- tinue reading data from the route’s StreamReader.

For routing service, domain routes, auto routes and auto topic routes, the execution of this com- mand will resume the contained topic routes and routes.

5.2.10save

save <target_routing_service>

This command writes the current configuration to a file. The file itself is specified with <save_path> (see page 2-14). If <save_path> has not been specified, the save command will fail. If the file specified by <save_path> already exists, the file will be overwritten.

The saved configuration is functionally equivalent to the loaded XML file plus any updates (either from an update command or other remote commands that change the configuration, such as add_peer). However it may not be textually equivalent. For example, the saved XML configuration may explicitly contain default values that were not in the initial XML.

5.2.11unload

unload <target_routing_service>

The unload command unloads the current configuration that the target_routing_service is using, so you can change it with a subsequent load (Section 5.2.7) command.

The target_routing_service must be disabled for this command to succeed.

5-5

5.2.12update

update <target_routing_service> [<entity_name>] [<xml_url>|<assignment_expr> [remote|local]

The update command changes the configuration of a specific entity. Table 5.1 shows the param- eters that can be changed for each entity.

Table 5.1 Changeable Parameters

Entity

Mutable

Immutable

(changeable any time)

(only changeable when entity is disabled)1

 

 

 

 

<monitoring>/<enabled>

<monitoring>/<statistics_sampling_period>

 

<monitoring>/<historical_statistics>

 

<monitoring>/<status_publication_period>

 

<monitoring>/<domain_id>

Routing

<entity_monitoring>/<enabled>

<entity_monitoring>/<statistics_sampling_period>

Service

<entity_monitoring>/<status_publication_period>

<entity_monitoring>/<historical_statistics>

 

<administration>/<save_path>

 

<administration>/

 

<administration>/<autosave_on_update>

 

<all except save_path and autosave_on_update>

 

 

 

 

 

 

<connection_x>: Mutable properties in

<connection_x>: Immutable properties in

 

<property> (adapter specific)

<property> (adapter specific).

Domain

<participant_x>: Mutable QoS policies in

<participant_qos>: Immutable QoS policies in

route

<participant_qos>

<participant_qos>

 

<entity_monitoring>/<enabled>

<entity_monitoring>/<statistics_sampling_period>

 

<entity_monitoring>/<status_publication_period>

<entity_monitoring>/<historical_statistics>

 

 

 

 

For non-Connext adapter: Mutable properties in

For non-Connext adapter: Mutable properties in

 

<property> (adapter specific)

<property> (adapter specific)

Session

For Connext adapter: Mutable QoS policies in

For Connext adapter: Immutable QoS policies in

<publisher_qos> and <subscriber_qos>

<publisher_qos> and <subscriber_qos>

 

 

<entity_monitoring>/<enabled>

<entity_monitoring>/<statistics_sampling_period>

 

<entity_monitoring>/<status_publication_period>

<entity_monitoring>/<historical_statistics>

 

 

 

 

Mutable properties in <property> (adapter specific)

Immutable properties in <property>

 

Mutable properties in

(adapter specific)

Route

Immutable properties in

<transformation>/<property>

 

<transformation>/<property>

 

(transformation specific)

 

(transformation specific)

 

 

 

 

 

 

Mutable QoS policies in <datawriter_qos> and

 

 

<datareader_qos>

 

 

Mutable properties in

 

 

<transformation>/<property>

Immutable QoS policies in <datawriter_qos> and

 

(transformation specific)

<datareader_qos>

Topic

<route_types>

<creation_mode>

Route

<propagate_dispose>

<content_filter>/<expression>

 

<propagate_unregister>

<entity_monitoring>/<statistics_sampling_period>

 

<publish_with_original_info>

<entity_monitoring>/<historical_statistics>

 

<content_filter>/<parameter>

 

 

<entity_monitoring>/<enabled>

 

 

<entity_monitoring>/<status_publication_period>

 

 

 

 

5-6

Table 5.1 Changeable Parameters

Entity

Mutable

Immutable

(changeable any time)

(only changeable when entity is disabled)1

 

 

 

Auto

Mutable properties in <property> (adapter specific)

Immutable properties in <property>

Route

 

(adapter specific)

 

 

Immutable QoS policies in <datawriter_qos> and

 

Mutable QoS policies in <datawriter_qos> and

<datareader_qos>

 

<datareader_qos>

<creation_mode>

Auto

<propagate_dispose>

<allow_topic_name_filter>

<propagate_unregister>

<allow_registered_type_name_filter>

Topic

<publish_with_original_info>

<deny_topic_name_filter>

Route

<content_filter>/<parameter>

<deny_registered_type_name_filter>

 

 

<entity_monitoring>/<enabled>

<content_filter>/<expression>

 

<entity_monitoring>/<status_publication_period>

<entity_monitoring>/<statistics_sampling_period>

 

 

<entity_monitoring>/<historical_statistics>

 

 

 

1. Monitoring parameters can also be changed when monitoring is disabled (even when the entity is enabled).

If you try to change an immutable parameter in an entity that is enabled, you will receive an error message. To change an immutable parameter, you must disable the routing service entity, change the parameter, and then enable the routing service entity again.

You can send an XML snippet (or an assignment expression) that only contains the values you want to change for that entity, or you can send a whole well-formed configuration file.

If you send an XML snippet (or an assignment expression), only the changes you specify will take effect.

For example, suppose you send this command:

update ShapeRouter DomainRoute1::Session1::SquareToCircles str://"<topic_route><input><datareader_qos><deadline><period> <sec>1</sec></period></deadline></datareader_qos></input></topic_route>"

or

update ShapeRouter DomainRoute1::Session1::SquareToCircles topic_route.input.datareader_qos.deadline.period.sec = 1

The topic route DomainRoute1::Session1::SquareToCircles will only change the period value in the Deadline QoS for that particular DataReader.

Now suppose that later on you send this command:

update ShapeRouter DomainRoute1::Session1::SquareToCircles str://"<topic_route><input><datareader_qos><property><value><element> <name>MyProp</name><value>MyValueRemote</value></element></value> </property><datareader_qos></input></topic_route>"

This would only change the Property QoS; the Deadline QoS would keep the setting from the prior command.

If you send a well-formed configuration file (starting with <dds><routing_service>), the properties in the route (QoS values in the topic route) will be completely replaced with the properties (QoS values) defined in the XML code. If a QoS value for a topic route is not defined in the XML code, Routing Service will use the Connext default.

5-7

5.3Accessing Routing Service from a Connext Application

You can create a DataWriter for the command topic to write Routing Service administration commands and optionally create a DataReader for the response topic to receive confirmations.

A more powerful and easier way is to use the Request-Reply API (only available with Connext Messaging). You can create a Requester for these topics that will write command requests and wait for confirmations.

The topics are:

rti/routing_service/administration/command_request

rti/routing_service/administration/command_response

The types are:

RTI::RoutingService::Administration::CommandRequest

RTI::RoutingService::Administration::CommandResponse

You can find the IDL definitions for these types in <Routing Service installation directory>/ resource/idl/RoutingServiceAdministration.idl.

The QoS configurations of yur DataWriter and DataReader, or your Requester (if you are using the Request-Reply API), must be compatible with the one used by the routing service (see how this is configured in Section 2.4.3).

When you send an XML string URL (str://"<xml_code>") with the load and update commands, if the string is longer than XML_URL_MAX_LENGTH (in the IDL file), you will have to split the string and send several samples, setting the "final" field to false in all but the last sample.

Likewise, the get command may generate a response longer than RESPONSE_MAX_LENGTH (in the IDL file) that will be received as several samples. You will have to concatenate the mes- sages from each one of the samples until a sample with the “final” field set to true is received. This sample is the last sample of the response.

Example 1:

The following example shows how to send a command to update the Deadline QoS policy for a topic route's DataReader:

/* Create entities: participant, publisher, topic, datawriter...*/
/* ... */

RTI_RoutingService_CommandRequest * cmdRequest =

RTI_RoutingService_CommandRequestTypeSupport::create_data();

/* By specifying a unique ID for this command, you can identify its response later on */

cmdRequest->id.host = /* host ID */;

cmdRequest->id.app = /* process ID */;

cmdRequest->id.invocation = ++invocationCounter;

/* Send this command to a routing service called MyRouter */

strcpy(cmdRequest->target_router, "MyRouter");

5-8

/* The command type is update */

cmdRequest->command._d = RTI_ROUTING_SERVICE_COMMAND_UPDATE;

/* Specify entity name to update and XML code to define a new configuration */

strcpy(cmdRequest->command._u.entity_desc.name, "DomainRoute1::Session1::TopicRoute1");

/* When we use an XML snippet, the first tag we specify is that of the entity, <topic_route> in this case */

strcpy(cmdRequest->command._u.entity_desc.xml_url.content, "str://\"<topic_route>\

<input>\ <datareader_qos>\ <deadline>\

<period>\

<sec>10</sec>\

</period>\

</deadline>\ </datareader_qos>\ </input>\ </topic_route>\"");

/* The content above is small enough to send it in one sample. Otherwise (if the length were > XML_URL_MAX_LENGTH) we would have to split it in multiple partial strings,

each < XML_URL_MAX_LENGTH, and set final = 0 for all the samples but the last one */

cmdRequest->command._u.entity_desc.xml_url.final = 1; RTI_RoutingService_CommandRequest_writer->write(cmdRequest, ...);

Example 2, Using the Request-Reply Communication Pattern:

This example uses the RTI Connext Messaging Request-Reply API1. This example shows a Java application that creates a Requester that can communicate with the Routing Service remote- administration server. It sends two requests (Routing Service remote commands) to disable and then enable Routing Service. Each request receives a reply with the result for that command. When using the Request-Reply API, Routing Service will efficiently deliver those replies only to the original Requester.

Note: In the command topic, the values for id.host and id.app are not relevant in this example, but they are still needed when using the regular Connext API.

import RTI.RoutingService.Administration.CommandKind;

import RTI.RoutingService.Administration.CommandRequest;

import RTI.RoutingService.Administration.CommandRequestTypeSupport;

1. The Request-Reply Communication Pattern is only available with RTI Connext Messaging. For information, see the

Core Libraries and Utilities User’s Manual or API Reference HTML documentation.

5-9

import RTI.RoutingService.Administration.CommandResponse;

import RTI.RoutingService.Administration.CommandResponseTypeSupport;

import com.rti.connext.infrastructure.Sample;

import com.rti.connext.infrastructure.WriteSample;

import com.rti.connext.requestreply.Requester;

import com.rti.connext.requestreply.RequesterParams;

import com.rti.dds.domain.DomainParticipant;

import com.rti.dds.domain.DomainParticipantFactory;

import com.rti.dds.infrastructure.Duration_t;

import com.rti.dds.infrastructure.InstanceHandleSeq;

import com.rti.dds.infrastructure.StatusKind;

import com.rti.dds.publication.DataWriterQos;

/**

* How to use Routing Service administration through a Requester */

public class CommandExample {

static final String COMMAND_TOPIC =
"rti/routing_service/administration/command_request";

static final String RESPONSE_TOPIC =
"rti/routing_service/administration/command_response";

private static final Duration_t MAX_WAIT = new Duration_t(10, 0);

public static void main(String[] args) throws InterruptedException {

//

// Create DomainParticipant

//

DomainParticipant participant = DomainParticipantFactory.get_instance() .create_participant(

55, DomainParticipantFactory.PARTICIPANT_QOS_DEFAULT, null, StatusKind.STATUS_MASK_NONE);

if (participant == null) {

throw new IllegalStateException("Participant creation failed");

}

try {

//

// Create requester for Routing Service remote-admin topics

//

Requester<CommandRequest, CommandResponse> requester = new Requester<CommandRequest, CommandResponse>(

new RequesterParams(participant, CommandRequestTypeSupport.get_instance(), CommandResponseTypeSupport.get_instance())

5-10

.setRequestTopicName(COMMAND_TOPIC)

.setReplyTopicName(RESPONSE_TOPIC));

DataWriterQos writerQos = new DataWriterQos();

requester.getRequestDataWriter().get_qos(writerQos);

System.out.println("rel" + writerQos.reliability.kind);

try { System.out.println(

"Waiting to discover Routing Service");

InstanceHandleSeq handles = new InstanceHandleSeq();

while (handles.isEmpty()) { requester.getRequestDataWriter().

get_matched_subscriptions(handles); Thread.sleep(200);

}

System.out.println("Matched subscription");

//

// Send DISABLE command

//

WriteSample<CommandRequest> request = requester .createRequestSample();

request.getData().id.host = 1;

request.getData().id.app = 1;

request.getData().id.invocation = 1;

request.getData().target_router = "TestRouter";

request.getData().command._d = CommandKind.RTI_ROUTING_SERVICE_COMMAND_DISABLE;

requester.sendRequest(request);

// Receive the reply

Sample<CommandResponse> reply = requester.createReplySample();

boolean received = requester.receiveReply(reply, MAX_WAIT);

if (!received) {

throw new IllegalStateException("Response not received");

}

System.out.println("Received response: " + reply.getData().message);

5-11

//

// Send ENABLE command

//

request.getData().id.invocation = 2;

request.getData().command._d = CommandKind.RTI_ROUTING_SERVICE_COMMAND_ENABLE;

requester.sendRequest(request);

// Receive the reply

received = requester.receiveReply(reply, MAX_WAIT);

if (!received) {

throw new IllegalStateException("Response not received");

}

System.out.println("Received response: " + reply.getData().message);

} finally { requester.close();

}

} finally { participant.delete_contained_entities();

DomainParticipantFactory.get_instance().delete_participant( participant);

}

}

}

5-12

Chapter 6 Monitoring Routing Service from a Remote Location

You can monitor Routing Service remotely by subscribing to special Connext topics. By subscribing to these topics, any Connext application can receive information about the configuration and operational status of Routing Service.

Being able to monitor the state of a Routing Service instance is an important tool that allows you to detect problems. For example, looking at the latency statistics for a route might show you that the performance of a transformation in the route is not as expected. Looking at the input samples per second in the different sessions, you might see that one session is receiving most of the traffic. In that case, you could reassign some of the routes to other sessions to improve load balancing.

Routing Service can publish status for the following kinds of entities:

1.Routing Service itself (<routing_service>)

2.Domain Route <domain_route>)

3.Session (<session>)

4.Route (<route> and <topic_route>)

5.Topic Route (<auto_route> and <auto_topic_route>)

For each of the above kinds of entities, Routing Service creates two topics:

rti/routing_service/monitoring/<tag>_data describes the entity’s configuration

rti/routing_service/monitoring/<tag>_status_set describes the entity’s operational status

With the corresponding types:

RTI::RoutingService::Monitoring::<tag>Data

RTI::RoutingService::Monitoring::<tag>StatusSet

Where <tag> is one of the following entity kind tags: RoutingService, DomainRoute, Session,

Route, or AutoRoute.

6-1

6.1Enabling Remote Monitoring

By default, remote monitoring is disabled in Routing Service for security and performance reasons.

To enable remote monitoring, you can use the <monitoring> tag (see Section 2.4.4) or the –remoteMonitoringDomainId command-line parameter, which enables remote monitoring and sets the domain ID for data publication. For more information about the command-line options, see Section 3.1 in the Getting Started Guide.

When remote monitoring is enabled, Routing Service creates:

1 DomainParticipant

1 Publisher

5 DataWriters for publishing configuration data (one for each kind of entity)

5 DataWriters for publishing status (one for each kind of entity).

The QoS values for these entities are described in Section 2.4.4.

6.2Monitoring Configuration Data

Configuration data for Routing Service entities is published in entity data topics. These topics are similar to the Connext builtin topics (DCPSParticipant, DCPSPublication, and DCPSSubscription) that provide information about the configuration of remote Connext entities.

This configuration data is published when:

An entity is created or enabled.

An entity is disabled or destroyed (a dispose message is published).

The entity’s configuration is modified using the remote command “update” (see Section 5.2.12).

The entity’s configuration is modified due to certain events in Routing Service. For exam- ple, discovery events may trigger the creation of StreamWriters and StreamReaders in a route.

The following sections describe the data available for each kind of Routing Service entity.

Configuration Data for Routing Service (Section 6.2.1)

Configuration Data for a Domain Route (Section 6.2.2)

Configuration Data for a Session (Section 6.2.3)

Configuration Data for a Route (Section 6.2.4)

Configuration Data for an Auto Route (Section 6.2.5)

Each section describes the IDL for the topics’ underlying data types. The IDL is also in the file

<Routing Service installation directory>/resource/idl/RoutingServiceMonitoring.idl.

6.2.1Configuration Data for Routing Service

The topic that publishes configuration data is called rti/routing_service/monitoring/ routing_service_data. This topic describes the configuration of the routing service but not its contained entities.

6-2

The IDL definition of the data type is:

struct RoutingServiceAdministrationData {

string<EXPRESSION_MAX_LENGTH> save_path;
boolean autosave_on_update;

};

struct RoutingServiceData {

string<ENTITY_NAME_MAX_LENGTH> name; //@key
string<ENTITY_NAME_MAX_LENGTH> group_name;
string<ENTITY_NAME_MAX_LENGTH> host_name;
long host_id;
long app_id;

RoutingServiceAdministrationData administration;

};

Table 6.1 on page 6-3 describes the members of the RoutingServiceData data type.

Table 6.1 RoutingServiceData

Field Name

Description

 

 

 

 

 

Key field.

 

Name of the routing service instance.

name

The name associated with the routing service instance can be assigned explicitly using

the command-line parameter –appName. If –appName is not used, the

 

<routing_service> tag name provided with -cfgName is used. If you use -

 

identifyExecution, the host name and process ID are appended to the name. For

 

example: RTI_RoutingService_myhost_1234

 

 

 

Name of the group to which the routing service belongs.

 

Routing services in the same group will not communicate with each other.

group_name

The group name is assigned using the attribute group_name in the <routing_service>

 

tag. If the attribute is not defined, the group name is automatically set to

 

RTI_RoutingService_<Host Name>_<Process ID>

 

 

host_name

Name of the host where the routing service is running.

 

 

host_id

Identifies the host where the routing service instance is running.

 

 

app_id

Process (task) ID of the routing service instance.

 

 

administration.

Specifies the file that will contain the saved configuration.

save_path

 

 

 

administration.

A boolean that, if true, automatically triggers a save command when configuration

auto_save_on_update

updates are received.

 

 

Routing Service data samples are published when:

The routing service instance is enabled.

The routing service instance is disabled (dispose sample).

Monitoring is enabled via remote administration.

6-3

6.2.2Configuration Data for a Domain Route

The topic that publishes domain route configuration data is called rti/routing_service/ monitoring/domain_route_data. The domain route data describes the configuration of the domain route and its connections but not its contained entities. Each connection can be defined with two different types, depending on if it is a Connext connection (<participant_1> or <participant_2) or a generic connection using an adapter (<connection_1 or <connection_2).

The IDL definition of the data type RTI::RoutingService::Monitoring::DomainRouteData is:

struct DomainRouteParticipantData {

long domain_id;
BuiltinTopicKey_t participant_key;

};

struct DomainRouteAdapterConnectionData {

string<ENTITY_NAME_MAX_LENGTH> plugin_name; sequence<Property, MAX_PROPERTIES> property;

};

union

DomainRouteConnectionData switch(AdapterKind) {

case RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND: DomainRouteParticipantData dds;

case RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND: DomainRouteAdapterConnectionData generic;

};

struct DomainRouteData {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

DomainRouteConnectionData connection_1;

DomainRouteConnectionData connection_2;

};

Table 6.2 describes the members of the DomainRouteData data type.

Table 6.2 DomainRouteData

Field Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned using –appName).

 

 

 

 

Key field

name

The domain route name. This is configured using the name attribute in the

 

<domain_route> tag.

 

 

6-4

Table 6.2 DomainRouteData

Field Name

Description

 

 

 

The configuration of the <connection_1> or <participant_1>.

connection_1

If it is a <connection_1>, the union discriminator is

RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if it is a <participant_1>,

 

 

the union discriminator is RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND

 

 

cconnection_1.dds.

Domain ID of the first domain route participant.

This domain ID is configured using the XML tag <domain_id> inside

domain_id

<participant_1>.

 

 

 

connection_1.dds.

Unique identifier for the first participant.

participant_key

 

 

 

connection_1.generic.

The name of the plugin used by the first connection (<connection_1>)

plugin_name

 

 

 

connection_1.generic.

The sequence of properties defined in the tag <property> inside <connection_1>

property

 

 

 

 

The configuration of the <connection_2> or <participant_2>.

connection_2

If it is a <connection_2>, the union discriminator is

RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if it is a <participant_2>,

 

 

the union discriminator is RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND

 

 

connection_2.dds.

Domain ID of the second domain route participant.

This domain ID is configured using the XML tag <domain_id> inside

domain_id

<participant_1>.

 

 

 

connection_2.dds.

Unique identifier for the second participant.

participant_key

 

 

 

connection_2.generic.

The name of the plugin used by the second connection (<connection_2>)

plugin_name

 

 

 

connection_2.generic.

The sequence of properties defined in the tag <property> inside <connection_2>

property

 

 

 

A domain route using Connext can be correlated with its corresponding participants using the fields connection_1.dds.participant_key and/or connection_2.dds.participant_key.

For example, let’s assume that we want to get the value of the PropertyQosPolicy associated with the first DomainParticipant of a domain route. To do that, we would subscribe to the participant Connext builtin topic and look for a sample where the key member is equal to participant1_key. From this sample, we can get the PropertyQosPolicy by accessing the member called property.

For additional information on how to subscribe to Connext builtin topics, see the RTI Core Libraries and Utilities User’s Manual.1

Domain-route data samples are published when:

The domain route is enabled.

The domain route is disabled (dispose sample).

Monitoring is enabled via remote administration.

6.2.3Configuration Data for a Session

The topic that publishes session configuration data is called rti/routing_service/monitoring/ session_data. The session data describes the configuration of the session but not its contained entities.

1.See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

6-5

The IDL definition of the data type is:

struct SessionData {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

long enabled_route_count;

sequence<Property, MAX_PROPERTIES> property;

};

Table 6.3 describes the fields in the SessionData data type.

Table 6.3 SessionData

Field Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned using –appName).

 

 

domain_route_name

Key field

The domain route name.

 

 

 

name

Key field

The session name, which is configured with the name attribute in the <session> tag.

 

 

 

enabled_route_count

The number of enabled routes.

 

 

property

The sequence of properties defined in the tag <property> inside <session>

 

 

Session data samples are published when:

The session is enabled.

The session is disabled (dispose sample).

An auto route/route inside the session is enabled.

An auto route/route inside the session is disabled.

Monitoring is enabled via remote administration.

6.2.4Configuration Data for a Route

The topic that publishes route configuration data is called rti/routing_service/monitoring/ route_data.

The IDL definition of the data type RTI::RoutingService::Monitoring::RouteData is:

struct TransformationData {

string<ENTITY_NAME_MAX_LENGTH> plugin_name; sequence<Property, MAX_PROPERTIES> property;

};

struct RouteAdapterData {

sequence<Property, MAX_PROPERTIES> property;

};

6-6

struct RouteDdsInputData {

long domain_id;

BuiltinTopicKey_t datareader_key;

string<EXPRESSION_MAX_LENGTH> content_filter_expression;

};

union

RouteInputAdapterData switch(AdapterKind) {

case RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND: RouteDdsInputData dds;

case RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND: RouteAdapterData generic;

};

struct RouteDdsOutputData {

long domain_id;

BuiltinTopicKey_t datawriter_key;

};

union RouteOutputAdapterData switch(AdapterKind) {

case RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND: RouteDdsOutputData dds;

case RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND: RouteAdapterData generic;

};

struct RouteInputData {

string<TOPIC_NAME_MAX_LENGTH> stream_name;
string<TYPE_NAME_MAX_LENGTH> registered_type_name;
long connection;

RouteCreationMode creation_mode;

DDSEntityState state;

RouteInputAdapterData adapter_data;

};

6-7

struct RouteOutputData {

string<TOPIC_NAME_MAX_LENGTH> stream_name;
string<TYPE_NAME_MAX_LENGTH> registered_type_name;
RouteCreationMode creation_mode;

DDSEntityState state;

RouteOutputAdapterData adapter_data;

};

struct RouteData {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name;//@key
string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key
string<ENTITY_NAME_MAX_LENGTH> session_name; //@key
string<ENTITY_NAME_MAX_LENGTH> name; //@key

string<ENTITY_NAME_MAX_LENGTH> auto_route_name;

boolean propagate_dispose;

boolean propagate_unregister;

boolean publish_with_original_info;

boolean publish_with_original_timestamp;

boolean route_types;

RouteInputData input;

RouteOutputData output;

sequence<TransformationData, MAX_TRANSFORMATIONS> transformations;

boolean paused;

};

Table 6.4 describes the fields in the RouteData topic data type.

Table 6.4 RouteData

Field Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned with –appName).

 

 

 

domain_route_name

Key field

The domain route name.

 

 

 

session_name

Key field

The session name.

 

 

 

 

Key field

name

The route name, which is configured using the name attribute in the <route> or

 

<topic_route> tag.

 

 

6-8

Table 6.4 RouteData

Field Name

Description

 

 

 

 

auto_route_name

If the route is contained in an auto-route, this field contains the auto-route name.

Otherwise, the field is initialized with the empty string.

 

 

 

 

(Connext topic routes only) Indicates if the topic route propagates

propagate_dispose

NOT_ALIVE_DISPOSE samples.

The propagation of NOT_ALIVE_DISPOSE samples is configured using the tag

 

 

<propagate_dispose> in <topic_route>.

 

 

 

(Connext topic routes only) Indicates if the topic route propagates

propagate_unregister

NOT_ALIVE_NO_WRITERS samples.

The propagation of NOT_ALIVE_NO_WRITERS samples is configured using the

 

 

tag <propagate_unregister> in <topic_route>.

 

 

 

(Connext topic routes only) Indicates if the topic route publishes the samples with

 

original writer info. Setting this option to true allows redundant topic routes and

publish_with_original_info

prevents the applications from receiving duplicate samples.

 

The publication with original writer info is configured using the tag

 

<publish_with_original_info> inside <topic_route>.

 

 

publish_with_original_

Indicates if the route is configured to publish the output samples with the same

timestamp

timestamp as that of the input sample.

 

 

 

Indicates if the input connection will use types discovered in the output

route_types

connection and viceversa for the creation of StreamWriters and StreamReaders.

The route types flag is configured using the tag <route_types> inside <route> or

 

 

<topic_route>.

 

 

input

The configuration of the route’s input, as contained in the tag <input> or

<dds_input> inside <route> or <topic_route>

 

 

 

 

Input stream name.

input.

The input stream name is configured using the tag <topic_name> inside

stream_name

<topic_route>/<input> or inside <route>/<dds_input> or the tag

 

<stream_name> inside <route>/<input>.

 

 

input.

Input registered name.

registered_type_

The input registered name is configured using the tag <registered_type_name>

name

inside <topic_route>/<input>, <route>/<dds_input> or <route>/<input>.

 

 

input.

Index of the input connection or participant (1 or 2).

The value of this field is used to determine whether the input of this route is the

connection

domain route’s connection 1/participant 1 or the connection 2/participant 2.

 

 

 

input.

Indicates when the StreamReader is created in the input.

creation_mode

The input creation mode is configured using the tag <creation_mode>.

 

 

input.

Indicates whether or not the StreamReader associated with a route is created.

state

 

 

 

 

Contains the configuration of the route’s input that is specific to either the

 

Connext adapter or a generic adapter.

input.

When a generic input is defined (<route>/<input>) then the union discriminator

adapter_data

is RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if it is a Connext

 

input (<topic_route>/<input> or <route>/<dds_input>), then the union

 

discriminator is RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND

 

 

input.

 

adapter_data.

(Connext input only) Domain ID of the input participant

dds.

 

domain_id

 

 

 

6-9

Table 6.4 RouteData

Field Name

Description

 

 

 

 

input.

(Connext input only) Unique identifier for the DataReader.

adapter_data.

The value of this field is meaningful only when the state is

dds.

RTI_ROUTING_SERVICE_CREATED_AND_ENABLED.

datareader_key

 

 

 

input.

(Connext input only) Content filter expression associated with the content filter

adapter_data.

for the topic route DataReader.

dds.

The expression is configured using the tag <content_filter>/<expression> inside

content_filter_

<topic_route>/<input> or <route>/<dds_input>

expression

 

 

 

input.

 

adapter_data.

(Not applicable for Connext input) The properties used to configure this route’s

generic.

StreamReader, specified with the tag <property> inside <route>/<input>

property

 

 

 

output

The configuration of the route’s output, as contained in the tag <output> or

<dds_output> inside <route> or <topic_route>

 

 

 

 

Output stream name.

output.

The output stream name is configured using the tag <topic_name> inside

stream_name

<topic_route>/<output> or inside <route>/<dds_output> or the tag

 

<stream_name> inside <route>/<output>.

 

 

output.

Output registered name.

registered_type_

The output registered name is configured using the tag <registered_type_name>

name

inside <topic_route>/<output>, <route>/<dds_output> or <route>/<output>.

 

 

output.

Indicates when the StreamWriter in created in the output.

creation_mode

The output creation mode is configured using the tag <creation_mode>.

 

 

output.

Indicates whether or not the StreamWriter associated with a route is created.

state

 

 

 

 

Contains the configuration of the route’s output that is specific to either the

 

Connext adapter or a generic adapter.

output.

When a generic output is defined (<route>/<output>) then the union

adapter_data

discriminator is RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if it is

 

a Connext output (<topic_route>/<output> or <route>/<dds_output>), then the

 

union discriminator is RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND

 

 

output.

 

adapter_data.

(Connext output only) Domain ID of the output participant

dds.

 

domain_id

 

 

 

output.

(Connext output only) Unique identifier for the DataWriter.

adapter_data.

The value of this field is only meaningful when datawriter_state is

dds.

RTI_ROUTING_SERVICE_CREATED_AND_ENABLED.

datawriter_key

 

 

 

output.

 

adapter_data.

(Not applicable for Connext output) The properties used to configure this route’s

generic.

StreamWriter, specified with the tag <property> inside <route>/<output>

property

 

 

 

6-10

Table 6.4 RouteData

Field Name

Description

 

 

 

 

 

List of transformations associated with a route.

 

For each transformation you will be able to retrieve the transformation plugin

transformations

name, and the properties.

Transformations are defined using the <transformation> tag inside <route> or

 

 

<topic_route>.

 

Note: in this version, only one transformation per route is supported.

 

 

paused

Indicates if a route or auto route has been paused with the remote command

pause.

 

 

 

The correlation between a route using Connext and its DataReader and DataWriter can be done using the fields datareader_key and datawriter_key.

For example, let’s assume that we want to retrieve the value of the DurabilityQosPolicy associated with the route’s DataWriter. To do that, we would subscribe to the publication Connext builtin topic and we would look for a sample where the key member is equal to datawriter_key. From this sample, we can get the DurabilityQosPolicy value accessing the member durability.

For additional information on how to subscribe to the Connext builtin topics, see the RTI Core Libraries and Utilities User’s Manual.1

Route data samples are published when:

The route is enabled.

The route is disabled (dispose sample).

The route configuration is modified using the remote command update.

The route’s StreamReader is created.

The route’s StreamReader is destroyed.

The route’s StreamWriter is created.

The route’s StreamWriter is destroyed.

Monitoring is enabled via remote administration.

6.2.5Configuration Data for an Auto Route

The topic that publishes auto route configuration data is called rti/routing_service/monitoring/ auto_route_data.

The IDL definition of the data type RTI::RoutingService::Monitoring::AutoRouteData is:

struct AutoRouteAdapterData {

sequence<Property, MAX_PROPERTIES> property;

};

struct AutoRouteDdsInputData {

long domain_id;

string<EXPRESSION_MAX_LENGTH> content_filter_expression;

};

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

6-11

union AutoRouteInputAdapterData switch(AdapterKind) {

case RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND: AutoRouteDdsInputData dds;

case RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND: AutoRouteAdapterData generic;

};

struct AutoRouteDdsOutputData {

long domain_id;

};

union AutoRouteOutputAdapterData switch(AdapterKind) {

case RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND: AutoRouteDdsOutputData dds;

case RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND: AutoRouteAdapterData generic;

};

struct AutoRouteInputData {

string<TOPIC_NAME_MAX_LENGTH> allow_stream_name_filter; string<TYPE_NAME_MAX_LENGTH> allow_registered_type_name_filter; string<TOPIC_NAME_MAX_LENGTH> deny_stream_name_filter; string<TYPE_NAME_MAX_LENGTH> deny_registered_type_name_filter;

long connection; RouteCreationMode creation_mode;

AutoRouteInputAdapterData adapter_data;

};

struct AutoRouteOutputData {

string<TOPIC_NAME_MAX_LENGTH> allow_stream_name_filter; string<TYPE_NAME_MAX_LENGTH> allow_registered_type_name_filter; string<TOPIC_NAME_MAX_LENGTH> deny_stream_name_filter; string<TYPE_NAME_MAX_LENGTH> deny_registered_type_name_filter; RouteCreationMode creation_mode;

AutoRouteOutputAdapterData adapter_data;

};

6-12

struct AutoRouteData {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key string<ENTITY_NAME_MAX_LENGTH> session_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

boolean propagate_dispose;

boolean propagate_unregister;

boolean publish_with_original_info;

boolean publish_with_original_timestamp;

long enabled_route_count; AutoRouteInputData input; AutoRouteOutputData output; boolean paused;

};

Table 6.5 describes the fields in the AutoRouteData data type.

Auto-topic-route data samples are published when:

The auto route is enabled.

The auto route is disabled (dispose sample).

The auto route configuration is modified using the remote command update.

A new route is created from the auto route

Monitoring is enabled via remote administration.

Table 6.5 AutoRouteData

Field Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned using –appName).

 

 

 

domain_route_name

Key field

The domain route name.

 

 

 

session_name

Key field

The session name.

 

 

 

 

Key field

name

The auto route name, which is configured using the name attribute in the

 

<auto_route> or <auto_topic_route> tags.

 

 

 

(Connext auto_topic routes only) Indicates if the topic route propagates

propagate_dispose

NOT_ALIVE_DISPOSE samples.

The propagation of NOT_ALIVE_DISPOSE samples is configured using the

 

 

tag <propagate_dispose> in <topic_route>.

 

 

 

(Connext auto_topic routes only) Indicates if the topic routes propagate

propagate_unregister

NOT_ALIVE_NO_WRITERS samples.

The propagation of NOT_ALIVE_NO_WRITERS samples is configured using

 

 

the tag <propagate_unregister> in <auto_topic_route>.

 

 

6-13

Table 6.5 AutoRouteData

Field Name

Description

 

 

 

 

 

(Connext auto_topic routes only) Indicates if the topic routes publish the

 

samples with original writer information. Setting this option to true allows

publish_with_original_info

redundant topic routes and prevents applications from receiving duplicate

samples.

 

 

The publication with original writer info is configured using the tag

 

<publish_with_original_info> inside <auto_topic_route>.

 

 

publish_with_original_

Indicates if the routes are configured to publish the output samples with the

timestamp

same timestamp as that of the input sample.

 

 

enabled_route_count

The number of enabled routes associated with the auto route.

 

 

input

The configuration of the auto route input, as contained in the tag <input> or

<dds_input> inside <auto_route> or <auto_topic_route>

 

 

 

 

Topics that do not pass this filter in the input participant will not trigger the

input.

creation of routes.

allow_stream_

This filter is configured using the tag <allow_topic_name_filter> inside

name_filter

<auto_topic_route>/<input> or inside <auto_route>/<dds_input> or the tag

 

<allow_stream_name_filter> inside <auto_route>/<input>

 

 

 

Topic with types that do not pass this filter in the input participant will not

input.

trigger the creation of routes.

allow_registered_

This filter is configured using the tag <allow_registered_type_name_filter>

type_name_filter

inside <auto_topic_route>/<input>, <auto_route>/<dds_input>, or

 

<auto_route>/<input>.

 

 

 

Topics that pass this filter in the input participant will not trigger the creation

input.

of routes.

deny_stream_

This filter is configured using the tag <deny_topic_name_filter> inside

name_filter

<auto_topic_route>/<input>. or inside <auto_route>/<dds_input> or the

 

tag <deny_stream_name_filter> inside <auto_route>/<input>.

 

 

 

Topics with types that pass this filter in the input participant will not trigger

input.

the creation of routes.

deny_registered_

The input deny registered type name filter is configured using the tag

type_name_filter

<deny_registered_type_name_filter> inside <auto_topic_route>/<input>,

 

<auto_route>/<dds_input>, or <auto_route>/<input>.

 

 

 

Index of the input connection or participant (1 or 2).

input.

The value of this field is used to determine whether the input of this auto

connection

route is the domain route’s connection 1/participant 1 or the connection 2/

 

participant 2.

 

 

input.

Indicates when the StreamReader is created in the input.

creation_mode

The input creation mode is configured using the tag <creation_mode>.

 

 

 

Contains the configuration of the auto route’s input that is specific to either

 

the Connext adapter or a generic adapter.

input.

When a generic input is defined (<auto_route>/<input>), the union

discriminator is RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if

adapter_data

it is a Connext input (<auto_topic_route>/<input> or <auto_route>/

 

 

<dds_input>), the union discriminator is

 

RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND.

 

 

input.

 

adapter_data.dds.

(Connext input only) Domain ID of the input participant

domain_id

 

 

 

input.

(Connext input only) Content filter expression associated with the content

filter for the topic route DataReader.

adapter_data.dds.

The expression is configured using the tag <content_filter>/<expression>

content_filter_expression

 

inside <topic_route>/<input> or <route>/<dds_input>

6-14

Table 6.5 AutoRouteData

Field Name

Description

 

 

 

 

input.

(Not applicable for Connext input) The properties used to configure this

adapter_data.

route’s StreamReader, specified with the tag <property> inside <route>/

generic.property

<input>

 

 

output

The configuration of the auto route output, as contained in the tag <output>

or <dds_output> inside <auto_route> or <auto_topic_route>

 

 

 

 

Topics that do not pass this filter in the output participant will not trigger the

output.

creation of routes.

allow_stream_

This filter is configured using the tag <allow_topic_name_filter> inside

name_filter

<auto_topic_route>/<output> or inside <auto_route>/<dds_output> or the

 

tag <allow_stream_name_filter> inside <auto_route>/<output>.

 

 

 

Topics with types that do not pass this filter in the output participant will not

output.

trigger the creation of routes.

allow_registered_

This filter is configured using the tag <allow_registered_type_name_filter>

type_name_filter

inside <auto_topic_route>/<output>, <auto_route>/<dds_ioutput>, or

 

<auto_route>/<output>.

 

 

 

Topics that pass this filter in the output participant will not trigger the

output.

creation of routes.

The output deny topic name filter is configured using the tag

deny_stream_

<deny_topic_name_filter> inside <auto_topic_route>/<output>. or inside

name_filter

<auto_route>/<dds_output> or the tag <deny_stream_name_filter> inside

 

 

<auto_route>/<output>.

 

 

 

Topics with types that pass this filter in the output participant will not trigger

output.

the creation of routes.

deny_registered_

The output deny registered type name filter is configured using the tag

type_name_filter

<deny_registered_type_name_filter> inside <auto_topic_route>/<output>,

 

<auto_route>/<dds_output> , or <auto_route>/<output>.

 

 

output.

Indicates when the StreamWriter in created in the output.

creation_mode

The output creation mode is configured using the tag <creation_mode>..

 

 

 

Contains the configuration of the auto_route’s output that is specific to either

 

the Connext adapter or a generic adapter.

output.

When a generic output is defined (<auto_route>/<output>), the union

discriminator is RTI_ROUTING_SERVICE_GENERIC_ADAPTER_KIND; if

adapter_data

it is a Connext output (<auto_topic_route>/<output> or <auto_route>/

 

 

<dds_output>), the union discriminator is

 

RTI_ROUTING_SERVICE_DDS_ADAPTER_KIND.

 

 

output.

 

adapter_data.dds.

(Connext output only) Domain ID of the output participant

domain_id

 

 

 

output.

(Not applicable for Connext output) The properties used to configure this

adapter_data.

route’s StreamWriter, specified with the tag <property> inside <route>/

generic.property

<output>

 

 

paused

Indicates if a route or auto route has been paused with the remote command

pause.

 

 

 

6.3Monitoring Status

Operational status for Routing Service entities is published in entity status_set topics. This information changes continuously and is computed and published periodically.

6-15

The status information for the different entities is composed primarily of statistics. Section 6.3.1 explains how these statistics are calculated and published. These sections describe the status information associated with each kind of entity:

Status Information for the Routing Service (Section 6.3.2)

Domain Route Status (Section 6.3.3)

Status Information for a Session (Section 6.3.4)

Status Information for a Route (Section 6.3.5)

Status Information for an Auto Route (Section 6.3.6)

Each section describes the IDL for the topics’ underlying data types. The IDL is also in the file

<Routing Service installation directory>/resource/idl/RoutingServiceMonitoring.idl.

6.3.1How the Statistics are Generated

6.3.1.1Statistics Publication

Routing Service reports multiple statistics as part of the different status sets. For example, for a route the status contains statistical metrics about the input and output samples per second (throughput).

struct RouteStatusSet {

...

StatisticVariable input_samples_per_s;

StatisticVariable output_samples_per_s;

...

};

The statistical information is published periodically in the form StatisticVariables.

The period at which statistics are published is configurable using the tag <status_publication_ period> (see Section 2.4.4).

For a given variable, Routing Service computes the metrics in StatisticMetrics during specific time frames.

struct StatisticMetrics {

unsigned long long period_ms;

long long count;

float mean;

float minimum;

float maximum;

float std_dev;

};

6-16

struct StatisticVariable {

StatisticMetric publication_period_metrics;

sequence

<StatisticMetrics, MAX_HISTORICAL_METRICS> historical_metrics;

};

The count is the sum of all the values received during the time frame. For example, in the case of input_sample_per_s and output_sample_p_s, count is the number of samples received during the time frame. For latency, count is the sum of all the latency times for the samples received during the time frame.

If status publication is enabled (see Section 2.4.4), Routing Service always publishes the statistics corresponding to the time between two status publications (publication_period_metrics). You can also select additional windows on a per entity basis using the tag <historical_statistics> (see Section 2.4.4). The sequence historical_metrics in StatisticVariable contains values corresponding to the windows that have been enabled:

5-sec. metrics correspond to activity in the last five seconds.

1-min. metrics correspond to activity in the last minute.

5-min. metrics correspond to activity in the last five minutes.

1-hour metrics correspond to activity in the last hour.

Up-time metrics correspond to activity since the entity was enabled.

Each window has a field called period_ms that identifies its size in milliseconds. For the publication_period_metrics, this field contains the publication period. For the up-time metrics, this field contains the time since the entity was enabled. For the other windows, this field contains a fixed value that identifies the window size (5000 for the 5-second window, 60000 for the one-minute window, etc).

6.3.1.2Statistics Calculation

The accuracy of the statistics calculation process is determined by the value of the statistics sampling period. This period specifies how often statistics are gathered and is configured on a per entity basis using the tag <statistics_sampling_ period> (see Section 2.4.4).

As a general rule, the statistics_sampling_period of an entity must be smaller than its status_publication_period. A small statistics_sampling_period provides more accurate statistics at expense of increasing the memory consumption and decreasing performance.

6.3.2Status Information for the Routing Service

The topic that publishes routing service status is called rti/routing_service/monitoring/ routing_service_status _set.

The IDL definition of the data type is:

6-17

struct RoutingServiceStatusSet {

string<ENTITY_NAME_MAX_LENGTH> name; //@key

StatisticVariable cpu_usage_percentage;

StatisticVariable physical_memory_kb;

StatisticVariable total_memory_kb;

long uptime;

StatisticVariable host_cpu_usage_percentage;

StatisticVariable host_free_memory_kb; unsigned long host_total_memory_kb;

StatisticVariable host_free_swap_memory_kb; unsigned long host_total_swap_memory_kb;

long host_uptime;

};

Table 6.6 describes the fields in the RoutingServiceStatusSet data type.

Table 6.6 RoutingServiceStatusSet

Field Name

Description

 

 

 

 

 

Key field

 

Name of the routing service instance.

 

The name associated with the Routing Service instance can be assigned

name

explicitly by using the –appName command-line parameter. If –appName is

not used, the <routing_service> tag name provided with -cfgName is used. If

 

 

you use the –identifyExecution command-line parameter , the host name and

 

the process ID are appended to the name. For example:

 

RTI_RoutingService_myhost_1234

 

 

 

Statistic variable that provides the percentage of CPU usage of the Routing

cpu_usage_percentage

Service process over different time windows.

 

This variable is only supported on Windows and Linux systems.

 

 

 

Statistic variable that provides the physical memory utilization of the Routing

physical_memory_kb

Service process.This variable is only supported on Windows and Linux

 

systems.

 

 

 

Statistic variable that provides the virtual memory utilization of the Routing

total_memory_kb

Service process.This variable is only supported on Windows and Linux

 

systems.

 

 

uptime

Contains the time elapsed since the Routing Service process started running.

This value is only supported on Windows and Linux systems.

 

 

 

 

Statistic variable that provides the global percentage of CPU usage on the host

host_cpu_usage_percentage

where Routing Service is running. This variable is only supported on Windows

 

and Linux systems.

 

 

 

Statistic variable that provides the amount of free physical memory on the host

host_free_memory_kb

where Routing Service is running. This variable is only supported on Windows

 

and Linux systems.

 

 

6-18

Table 6.6 RoutingServiceStatusSet

Field Name

Description

 

 

 

 

host_total_memory_kb

Contains the total memory of the host where Routing Service is running. This

variable is only supported on Linux systems.

 

 

 

 

Statistic variable that provides the amount of free swap memory on the host

host_free_swap_memory_kb

where Routing Service is running. This value is only supported on Linux

 

systems.

 

 

host_total_swap_memory_kb

Contains the total swap memory of the host on which Routing Service is

 

running. This value is only supported on Linux systems.

host_uptime

Contains the time elapsed since the host on which Routing Service is running

started running. This value is only supported on Windows and Linux systems.

 

 

 

6.3.3Domain Route Status

The topic that publishes domain route status is called rti/routing_service/monitoring/ domain_route_status_set.

The domain route status aggregates the statistics of the routes contained in it: the mean of the means in the routes, the absolute maximum and minimum across routes, the mean of the standard deviation and the total count.

The IDL definition of the data type is:

struct DomainRouteStatusSet {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

StatisticVariable input_samples_per_s;

StatisticVariable input_bytes_per_s;

StatisticVariable output_samples_per_s;

StatisticVariable output_bytes_per_s;

StatisticVariable latency_s;

};

Table 6.7 describes the fields in the DomainRouteStatusSet data type.

Table 6.7 DomainRouteStatusSet

Field Name

Description

 

 

 

 

routing_service_name

Key field

 

The routing service name (assigned with –appName).

 

Key field

name

The domain route name, configured using the name attribute in the <domain_route>

 

tag.

 

 

 

Statistic variable that provides information about the input samples per second across

input_samples_per_s

routes.

Input samples refer to the samples that are taken by the sessions from the routes’s

 

 

StreamReaders.

 

 

 

Statistic variable that provides information about the input bytes per second across

input_bytes_per_s1

routes. Input bytes refer to the bytes that are taken by the sessions from the routes’s

StreamReaders. These bytes only refer to the serialized samples. The protocol headers

 

 

(UDP, RTPS) are not included.

 

 

6-19

Table 6.7 DomainRouteStatusSet

Field Name

Description

 

 

 

 

 

Statistic variable that provides information about the output samples per second

output_samples_per_s

across routes.

Output samples refer to the samples that are published out by the session threads

 

using the route’s StreamWriters.

 

 

 

Statistic variable that provides information about the output bytes per second across

 

routes.

output_bytes_per_s

Output bytes refer to the bytes that are published out by the session threads using the

 

route’s StreamWriters. The variable only considers the bytes of the serialized samples.

 

Protocol headers (UDP, RTPS) are not included.

 

 

 

Statistic variable that provides information about the latency in seconds across routes.

latency_s

The latency in a route refers to the time elapsed between the sample read and write.

 

This is a good metric to monitor the health and performance of transformations.

 

 

1. The throughput measured in bytes can only be computed if the samples are DynamicData samples. If not, only the throughput measured in samples per second is available. This statement applies to all the statistic variables described in this chapter that measure throughput in bytes per second.

6.3.4Status Information for a Session

The topic that publishes session status is called rti/routing_service/monitoring/ session_status_set.

The session status aggregates the statistics of the routes contained in it: the mean of the means in the routes, the absolute maximum and minimum across routes, the mean of the standard deviation and the total count.

The IDL definition of the data type is:

struct SessionStatusSet {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

StatisticVariable input_samples_per_s;

StatisticVariable input_bytes_per_s;

StatisticVariable output_samples_per_s;

StatisticVariable output_bytes_per_s;

StatisticVariable latency_s;

};

Table 6.8 describes the fields in the SessionStatusSet data type.

Table 6.8 SessionStatusSet

Field Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned with –appName).

 

 

domain_route_name

Key field

The domain route name

 

 

 

 

Key field

name

The session name.

 

The domain route name is configured using the name attribute in the <session> tag.

 

 

6-20

Table 6.8 SessionStatusSet

Field Name

Description

 

 

 

 

 

Statistic variable that provides information about the input samples per second across

input_samples_per_s

routes.

Input samples refer to the samples that are taken by the session from the routes’s

 

 

StreamReaders.

 

 

 

Statistic variable that provides information about the input bytes per second across

 

routes.

input_bytes_per_s

Input bytes refer to the bytes that are taken by the sessions from the routes’s

StreamReaders.

 

 

These bytes only refer to the serialized samples. The protocol headers (UDP, RTPS) are

 

not included.

 

 

 

Statistic variable that provides information about the output samples per second across

output_samples_per_s

routes.

 

Output samples refer to the samples that are published out by the session thread using

 

the route’s StreamWriters.

 

 

 

Statistic variable that provides information about the output bytes per second across

 

routes.

output_bytes_per_s

Output bytes refer to the bytes that are published out by the session thread using the

 

route’s StreamWriters. The variable only considers the bytes of the serialized samples.

 

Protocol headers (UDP, RTPS) are not included.

 

 

 

Statistic variable that provides information about the latency in seconds across routes.

latency_s

The latency in a route refers to the time elapsed between the sample read and write.

 

This is a good metric to monitor the health and performance of transformations.

 

 

6.3.5Status Information for a Route

The topic that publishes route status is called rti/routing_service/monitoring/route_status_set.

The IDL definition of the data type is:

struct RouteStatusSet {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key string<ENTITY_NAME_MAX_LENGTH> session_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

StatisticVariable input_samples_per_s;

StatisticVariable input_bytes_per_s;

StatisticVariable output_samples_per_s;

StatisticVariable output_bytes_per_s;

StatisticVariable latency_s;

};

6-21

Table 6.9 describes the fields in the RouteStatusSet data type.

Table 6.9 RouteStatusSet

Member Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned with -appName).

 

 

 

domain_route_name

Key field

The domain route name

 

 

 

session_name

Key field

The session name.

 

 

 

 

Key field

name

The route name.

The route name is configured using the name attribute in the <topic_route> or

 

 

<route> tags.

 

 

 

Statistic variable that provides information about the input samples per second in the

input_samples_per_s

route.

Input samples refer to the samples that are taken by the session from the route’s

 

 

StreamReader.

 

 

 

Statistic variable that provides information about the input bytes per second in the route.

 

Input bytes refer to the bytes that are taken by the session from the route’s

input_bytes_per_s

StreamReader.

 

These bytes only refer to the serialized samples. The protocol headers (UDP, RTPS) are

 

not included.

 

 

 

Statistic variable that provides information about the output samples per second in the

output_samples_per_s

routes.

Output samples refer to the samples that are published out by the session thread using

 

 

the route’s StreamWriters.

 

 

 

Statistic variable that provides information about the output bytes per second in routes.

output_bytes_per_s

Output bytes refer to the bytes that are published out by the session thread using the

route’s StreamWriter. The variable only considers the bytes of the serialized samples.

 

 

Protocol headers (UDP, RTPS) are not included.

 

 

 

Statistic variable that provides information about the latency in seconds in the routes.

latency_s

The latency in a route refers to the time elapsed between the sample read and write. This

 

is a good metric to monitor the health and performance of transformations.

 

 

6.3.6Status Information for an Auto Route

The topic that publishes auto route status is called rti/routing_service/monitoring/ route_status_set.

The auto route status aggregates the statistics of the routes created from it: the mean of the means in the routes, the absolute maximum and minimum across routes, the mean of the standard deviation and the total count.

6-22

The IDL definition of the data type is:

struct AutoRouteStatusSet {

string<ENTITY_NAME_MAX_LENGTH> routing_service_name; //@key string<ENTITY_NAME_MAX_LENGTH> domain_route_name; //@key string<ENTITY_NAME_MAX_LENGTH> session_name; //@key string<ENTITY_NAME_MAX_LENGTH> name; //@key

StatisticVariable input_samples_per_s;

StatisticVariable input_bytes_per_s;

StatisticVariable output_samples_per_s;

StatisticVariable output_bytes_per_s;

StatisticVariable latency_s;

};

Table 6.10 describes the fields in the AutoRouteStatusSet data type.

Table 6.10 AutoRouteStatusSet

Member Name

Description

 

 

 

 

routing_service_name

Key field

The routing service name (assigned with –appname).

 

 

domain_route_name

Key field

The domain route name.

 

 

 

session_name

Key field

The session name.

 

 

 

 

Key field

name

The auto route name.

The auto route name is configured using the name attribute in the <auto_topic_route>

 

 

or <auto_route> tags.

 

 

 

Statistic variable that provides information about the input samples per second across

input_samples_per_s

routes.

Input samples refer to the samples that are taken by the session from the auto routes’s

 

 

StreamReaders.

 

 

 

Statistic variable that provides information about the input bytes per second across

 

routes.

input_bytes_per_s

Input bytes refer to the bytes that are taken by the session from the auto routes’s

StreamReaders.

 

 

These bytes only refer to the serialized samples. The protocol headers (UDP, RTPS) are

 

not included.

 

 

 

Statistic variable that provides information about the output samples per second across

output_samples_per_s

routes.

Output samples refer to the samples that are published out by the session thread using

 

the auto route’s StreamWriters.

 

 

 

Statistic variable that provides information about the output bytes per second across

 

routes.

output_bytes_per_s

Output bytes refer to the bytes that are published out by the session thread using the

 

auto route’s StreamWriters. The variable only considers the bytes of the serialized

 

samples. Protocol headers (UDP, RTPS) are not included.

 

 

 

Statistic variable that provides information about the latency in seconds across routes.

latency_s

The latency in a route refers to the time elapsed between the sample read and write.

 

This is a good metric to monitor the health and performance of transformations.

 

 

6-23

Chapter 7 Traversing Wide Area Networks

Many systems today already rely on Connext to distribute their information across a Local Area Network (LAN). However, more and more of these systems are being integrated in Wide Area Networks (WANs). With Routing Service, you can scale Connext real-time publish/subscribe data-distribution beyond the current local networks and make it available throughout a WAN.

Out of the box, Routing Service only uses UDPv4 and Shared Memory transports to communicate with other Routing Services and Connext applications. This configuration is appropriate for systems running within a single LAN. However, using UDPv4 introduces several problems when trying to communicate with Connext applications running in different LANs:

UDPv4 traffic is usually filtered out by the LAN firewalls for security reasons.

Forwarded ports are usually TCP ports.

Each LAN may run in its own private IP address space and use NAT (Network Address Translation) to communicate with other networks.

To overcome these issues, Routing Service is distributed with a TCP transport that is NAT friendly. The transport can be configured via XML using the PropertyQosPolicy of the Routing Service’s participants. Figure 7.1 shows a typical scenario where two Routing Services are used to bridge two Connext applications running in two different LANs.

Figure 7.1 WAN Communication Using TCP Transport

Connext application 1 (LAN1)

Routing Service (LAN 1)

Participant 1

Participant 2

UDPv4

TCPv4

Transport

Transport

Connext application 2 (LAN2)

Routing Service (LAN 2)

Participant 1

Participant 2

TCPv4

UDPv4

Transport

Transport

Firewall/NAT

Firewall/NAT

Router

Router

 

TCP traffic only

7-1

The next sections explain how to use and configure the TCP transport with Routing Service.

7.1TCP Communication Scenarios

The TCP transport distributed with Routing Service can be used to address multiple communication scenarios that go from simple communication within a single LAN to complex communication scenarios across LANs where NATs and firewalls may be involved.

7.1.1Communication Within a Single LAN

TCP transport can be used as an alternative to UDPv4 to communicate Connext applications running inside the same LAN.

Figure 7.2 shows how to configure the TCP transport in this scenario.

parent.classid, transport_mode and server_bind_port are transport properties configured using the PropertyQosPolicy of the participant.

Initial Peers represents the peers to which the participant will be announced to. Usually, these peers are configured using the DiscoveryQosPolicy of the participant or the environment variable NDDS_DISCOVERY_PEERS. For information on the format of initial peers, see Section 7.2.1.

Figure 7.2 Communication within a Single LAN

7.1.2Symmetric Communication Across NATs

In NAT communication scenarios, each one of the LANs has a private IP address space. The communication with other LANs is done through NAT routers that translate private IP addresses and ports into public IP addresses and ports.

In symmetric communication scenarios, any instance of Routing Service can initiate TCP connections with other routing services. Figure 7.3 shows how to configure the TCP transport in this scenario.

Notice that initial peers refer to the public address of the Routing Service instances and not the LAN address. In addition, the transport associated with a Routing Service instance will have to be configured with its public_address so that this information can be propagated as part of the discovery process.

Because the public address and port of the Routing Service instances must be known before the communication is established, the NAT Routers will have to be configured statically to translate (forward) the private server_bind_port into a public port. This process is known as “static

7-2

NAT“ or “port forwarding” and it allows traffic originating in outer networks to reach designated peers in the LAN behind the NAT router.

Figure 7.3 Symmetric Communication across NATs

7.1.3Asymmetric Communication Across NATs

This scenario is similar to the previous one, except in this case the TCP connections can be initiated only by the Routing Service instance in LAN1. For security reasons, incoming connections to LAN1 are not allowed. Figure 7.4 shows how to configure the TCP transport in this scenario.

7.1.4Secure Communication

Security can be added on top of any of the above scenarios. You can have secure communication within the same LAN or across NATs.

To enable secure communication, modify the previous configurations as follows:

Change the transport class ID property (parent.classid) to be one of the following values: NDDS_TRANSPORT_CLASSID_TLSV4_LAN NDDS_TRANSPORT_CLASSID_TLSV4_WAN

Set at least a certificate of authority (through either the tls.verify.ca_file or tls.verify.ca_path properties), and the certificate identity (through either the tls.identity.certificate_chain, or tls.identity.certificate_chain_file properties)

Make sure to use ‘tlsv4_lan’ or ‘tlsv4_wan’ in the initial peers list as the prefix for all destination addresses.

To see the differences between a WAN scenario and the same scenario with TLS enabled, you can compare the two example configuration files:

shapes/tcp_transport.xml

shapes/tcp_transport_tls.xml

7-3

Figure 7.4 Asymmetric Communication Across NATs

Notice that the Routing Service on LAN 1 now does not have a public_address set (and its server_bind_port is set to zero), meaning that it cannot be reached from the outside network.

7.2Configuring the TCP Transport

The TCP transport is distributed as a shared library in <Routing Service installation directory>/ bin/<architecture>. The library is called nddstransporttcp.dll on Windows and libnddstransporttcp.so on UNIX-based systems.

For an example on how to use and configure the TCP transport with Routing Service see Example 8 - Using the TCP Transport with Routing Service (Section 4.8) in the Getting Started Guide.

As seen in the example, you can configure the properties of the transport in the XML configuration file using the appropriate name/value pairs in the DomainParticipant’s PropertyQoSPolicy. This will cause Routing Service to dynamically load the TCP transport library at run time and then implicitly create and register the transport plugin with Connext.

7.2.1TCP Transport Initial Peers

With the TCP transport, the addresses of the initial peers (NDDS_DISCOVERY_PEERS) that will be contacted during the discovery process have the following format:

For WAN communication: tcpv4_wan://<IP address or hostname>:<port>

For LAN communication: tcpv4_lan://<IP address or hostname>:<port>

For WAN+TLS communication: tlsv4_wan://<IP address or hostname>:port

For LAN+TLS communication: tlsv4_lan://<IP address or hostname>:port

7-4

For example:

setenv NDDS_DISCOVERY_PEERS tcpv4_wan://10.10.1.165:7400,tcpv4_wan://10.10.1.111:7400,tcpv4_lan://192.168.1.1:7500

When the TCP transport is configured for LAN communication (with the parent.classid property), the IP address is the LAN address of the peer and the port is the server port used by the transport (the server_bind_port property).

When the TCP transport is configured for WAN communication (with the parent.classid property), the IP address is the WAN or public address of the peer and the port is the public port that is used to forward traffic to the server port in the TCP transport.

When TLS is enabled, the transport settings are similar to WAN and LAN over TCP.

Figure 7.5 Initial Peers in WAN Communication

7.2.2Setting Up the TCP Transport Properties with the PropertyQoSPolicy

The PropertyQosPolicy allows you to set up name/value pairs of data and attach them to an entity, such as a DomainParticipant. The configuration of the TCP transport with Routing Service is done using the PropertyQosPolicy of the Domain Participants that are going to use the transport.

For a list of the properties that you can set for the TCP transport, see the RTI Core Libraries and Utilities User’s Manual.1

In the following example, participant_1 will communicate with other participants on the same LAN using UDP and Shared Memory transports; participant_2 will communicate with other participants in different LANs using the TCP transport.

<dds>

<routing_service name=”MyRoutingService”> <domain_route name=”MyDomainRoute”>

<participant_1> <domain_id>56</domain_id>

</participant_1>

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

7-5

<participant_2> <domain_id>57</domain_id> <participant_qos>

<transport_builtin> <mask>MASK_NONE</mask>

</transport_builtin> <property>

<value>

<element> <name>dds.transport.load_plugins</name> <value>dds.transport.TCPv4.tcp1</value> </element>

<element>

<name>dds.transport.TCPv4.tcp1.library</name>

<value>libnddstransporttcp.so</value>

</element>

<element> <name>dds.transport.TCPv4.tcp1.create_function </name> <value>NDDS_Transport_TCPv4_create</value> </element>

<element>

<name>dds.transport.TCPv4.tcp1.parent.classid

</name> <value>NDDS_TRANSPORT_CLASSID_TCPV4_WAN </value>

</element>

<element> <name>dds.transport.TCPv4.tcp1.public_address </name>

<value>18.181.0.31:8400</value>

</element>

<element> <name>dds.transport.TCPv4.tcp1.server_bind_port </name>

<value>7400</value>

</element>

</value>

</property> </participant_qos> </participant_2>

</domain_route> </routing_service>

</dds>

7-6

7.2.3TCP/TLS Transport Properties

Table 7.1 describes the TCP and TLS transport properties.

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

(prefix with Description ‘dds.transport.TCPv4.

tcp1.’)1

Required

dds.transport.

load_plugins Comma-separated strings indicating the prefix names of all plugins that will be

loaded by Connext. For example: “dds.transport.TCPv4.tcp1". You will use this (Note: this does not take a string as the prefix to the property names. See Footnote 1 on page 7-14.

prefix)

Note: you can load up to 8 plugins.

 

Required Must be "nddstransporttcp".

library

This library needs to be in the path during run time (in the LD_LIBRARY_PATH

 

environment variable on UNIX systems, in PATH for Windows systems).

 

 

create_function

Required Must be “NDDS_Transport_TCPv4_create”.

 

 

 

Used to register the transport plugin returned by

 

NDDS_Transport_TCPv4_create() (as specified by <TCP_prefix>.create_function)

aliases

to the DomainParticipant. Aliases should be specified as a comma-separated string,

with each comma delimiting an alias.

 

 

Default: the transport prefix (see Footnote 1)

 

Must be set to one of the following values:

 

NDDS_TRANSPORT_CLASSID_TCPV4_LAN

 

for TCP communication within a LAN

 

NDDS_TRANSPORT_CLASSID_TLSV4_LAN

parent.classid

for TLS communication within a LAN

NDDS_TRANSPORT_CLASSID_TCPV4_WAN

 

 

for TCP communication across LANs and firewalls

 

NDDS_TRANSPORT_CLASSID_TLSV4_WAN

 

for TLS communication across LAN and firewalls

 

Default: NDDS_TRANSPORT_CLASSID_TCPV4_LAN

 

 

 

Specifies the maximum number of buffers that Connext can pass to the send()

 

function of the transport plugin.

 

The transport plugin send() operation supports a gather-send concept, where the

 

send() call can take several discontiguous buffers, assemble and send them in a

 

single message. This enables Connext to send a message from parts obtained from

 

different sources without first having to copy the parts into a single contiguous

parent.gather_send_

buffer.

However, most transports that support a gather-send concept have an upper limit

buffer_count_max

on the number of buffers that can be gathered and sent. Setting this value will

 

 

prevent Connext from trying to gather too many buffers into a send call for the

 

transport plugin.

 

Connext requires all transport-plugin implementations to support a gather-send of

 

least a minimum number of buffers. This minimum number is defined as

 

NDDS_TRANSPORT_PROPERTY_GATHER_SEND_BUFFER_COUNT_MIN.

 

Default: 128

7-7

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

The maximum size of a message, in bytes, that can be sent or received by the

 

transport plugin.

parent.

If you set this higher than the default, the DomainParticipant’s buffer_size (in the

message_size_max

RECEIVER_POOL QosPolicy, see the RTI Core Libraries and Utilities User’s Manual2)

 

should also be changed.

 

Default: 9216

 

 

 

A list of strings, each identifying a range of interface addresses that can be used by

 

the transport.

 

Interfaces must be specified as comma-separated strings, with each comma

parent.

delimiting an interface.

 

allow_interfaces_list

For example: 10.10.*, 10.15.*

 

If the list is non-empty, this "white" list is applied before parent.

 

deny_interfaces_list.

 

Default: All available interfaces are used.

 

 

 

A list of strings, each identifying a range of interface addresses that will not be used

 

by the transport.

 

If the list is non-empty, deny the use of these interfaces.

parent.

Interfaces must be specified as comma-separated strings, with each comma

delimiting an interface.

deny_interfaces_list

For example: 10.10.*

 

 

This "black" list is applied after parent. allow_interfaces_list and filters out the

 

interfaces that should not be used.

 

Default: No interfaces are denied

 

 

 

Size, in bytes, of the send buffer of a socket used for sending. On most operating

 

systems, setsockopt() will be called to set the SENDBUF to the value of this

 

parameter.

send_socket_buffer_size

This value must be greater than or equal to parent. message_size_max

or -1.

 

 

The maximum value is operating system-dependent.

 

Default: -1 (setsockopt() (or equivalent) will not be called to size the send buffer of

 

the socket)

 

 

 

Size, in bytes, of the receive buffer of a socket used for receiving.

 

On most operating systems, setsockopt() will be called to set the RECVBUF to the

 

value of this parameter.

recv_socket_buffer_size

This value must be greater than or equal to parent. message_size_max

 

or -1. The maximum value is operating-system dependent.

 

Default: -1 (setsockopt() (or equivalent) will not be called to size the receive buffer

 

of the socket)

 

 

7-8

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

Prevents the transport plugin from using the IP loopback interface.

 

This property is ignored when parent.classid is

 

NDDS_TRANSPORT_CLASSID_TCPV4_WAN or

 

NDDS_TRANSPORT_CLASSID_TLSV4_WAN.

 

Two values are allowed:

ignore_loopback_

0: Enable local traffic via this plugin. The plugin will use and report the IP

loopback interface only if there are no other network interfaces (NICs) up on the

interface

system.

 

 

1: Disable local traffic via this plugin. This means “do not use the IP loopback

 

interface, even if no NICs are discovered.” This setting is useful when you want

 

applications running on the same node to use a more efficient plugin like shared

 

memory instead of the IP loopback.

 

Default: 1

 

 

 

Prevents the transport plugin from using a network interface that is not reported as

 

RUNNING by the operating system.

 

The transport checks the flags reported by the operating system for each network

 

interface upon initialization. An interface which is not reported as UP will not be

 

used. This property allows the same check to be extended to the IFF_RUNNING

 

flag implemented by some operating systems. The RUNNING flag means that "all

 

resources are allocated" and may be off if no link is detected (e.g., the network cable

ignore_nonrunning_

is unplugged).

 

interfaces

Two values are allowed:

 

0: Do not check the RUNNING flag when enumerating interfaces, just make sure

 

the interface is UP.

 

1: Check the flag when enumerating interfaces, and ignore those that are not

 

reported as RUNNING. This can be used on some operating systems to cause

 

the transport to ignore interfaces that are enabled but not connected to the

 

network.

 

Default: 1

 

 

 

Mask for the transport priority field. This is used in conjunction with

 

transport_priority_ mapping_low/transport_priority_ mapping_high to define the

 

mapping from Connext transport priority to the IPv4 TOS field. Defines a

 

contiguous region of bits in the 32-bit transport priority value that is used to

transport_priority_mask

generate values for the IPv4 TOS field on an outgoing socket.

For example, the value 0x0000ff00 causes bits 9-16 (8 bits) to be used in the mapping.

 

 

The value will be scaled from the mask range (0x0000 -0xff00 in this case) to the

 

range specified by low and high.

 

If the mask is set to zero, then the transport will not set IPv4 TOS for send sockets.

 

Default: 0

 

 

transport_priority_

Sets the low and high values of the output range to IPv4 TOS.

mapping_low

These values are used in conjunction with transport_priority_mask to define the

 

mapping from Connext transport priority to the IPv4 TOS field. Defines the low and

transport_priority_

high values of the output range for scaling.

Note that IPv4 TOS is generally an 8-bit value.

mapping_high

Default transport_priority_mapping_low: 0

 

 

Default transport_priority_mapping_high: 0xFF

 

 

server_socket_backlog

Determines the maximum length of the queue of pending connections.

Default: 5

 

 

 

7-9

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

Required for WAN communication

 

Public IP address and port (WAN address and port) associated with the transport

 

instantiation.The address and port must be separated with ‘:’.

 

For example: 10.10.9.10:4567

 

This field is only used when parent.classid is

 

NDDS_TRANSPORT_CLASSID_TCPV4_WAN or

 

NDDS_TRANSPORT_CLASSID_TLSV4_WAN.

public_address

The public address and port are necessary to support communication over a WAN

 

that involves Network Address Translators (NATs). Typically, the address is the

 

public address of the IP router that provides access to the WAN. The port is the IP

 

router port that is used to reach the private server_bind_port inside the LAN from

 

the outside. This value is expressed as a string in the form: ip[:port], where ip

 

represents the IPv4 address and port is the external port number of the router.

 

Note that host names are not allowed in the public_address because they may

 

resolve to an internet address that is not what you want (i.e., ‘localhost’ may map to

 

your local IP or to 127.0.0.1).

 

 

 

Private IP port (inside the LAN) used by the transport to accept TCP connections.

 

If this property is set to zero, the transport will disable the internal server socket,

 

making it impossible for external peers to connect to this node. In this case, the node

server_bind_port

is considered unreachable and will communicate only using the asynchronous

mode with other (reachable) peers.

 

 

For WAN communication, this port must be forwarded to a public port in the NAT-

 

enabled router that connects to the outer network.

 

Default: 7400

 

 

 

Allocation settings applied to read buffers.

 

These settings configure the initial number of buffers, the maximum number of

 

buffers and the buffers to be allocated when more buffers are needed.

read_buffer_allocation

Default:

read_buffer_allocation.initial_count = 2

 

 

read_buffer_allocation.max_count = -1 (unlimited)

 

read_buffer_allocation.incremental_count = -1 (number of buffers will keep

 

doubling on each allocation until it reaches max_count)

 

 

 

Allocation settings applied to buffers used for an asynchronous (non-blocking)

 

write.

 

These settings configure the initial number of buffers, the maximum number of

 

buffers, and the buffers to be allocated when more buffers are needed.

Default:

write_buffer_allocation.initial_count = 4

write_buffer_allocation.max_count = 1000

write_buffer_allocation

write_buffer_allocation.incremental_count = 10

Note that for the write buffer pool, the max_count is not set to unlimited. This is to avoid having a fast writer quickly exhaust all the available system memory, in case of a temporary network slowdown. When this write buffer pool reaches the maximum, the low-level send command of the transport will fail; at that point Connext will take the appropriate action (retry to send or drop it), according to the application’s QoS (if the transport is used for reliable communication, the data will still be sent eventually).

7-10

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

Allocation settings applied to buffers used to serialize and send control messages.

 

These settings configure the initial number of buffers, the maximum number of

 

buffers, and the buffers to be allocated when more buffers are needed.

control_buffer_allocation

Default:

control_buffer_allocation.initial_count = 2

 

 

control_buffer_allocation.max_count = -1 (unlimited)

 

control_buffer_allocation.incremental_count = -1 (number of buffers will

 

keep doubling on each allocation until it reaches max_count)

 

 

 

Allocation settings applied to control messages.

 

These settings configure the initial number of messages, the maximum number of

 

messages, and the messages to be allocated when more messages are needed.

control_message_

Default:

allocation

control_message_allocation.initial_count = 2

 

control_message_allocation.max_count = -1 (unlimited)

 

control_message_allocation.incremental_count = -1 (number of messages

 

will keep doubling on each allocation until it reaches max_count)

 

 

 

Allocation settings applied to control messages attributes.

 

These settings configure the initial number of attributes, the maximum number of

 

attributes, and the attributes to be allocated when more attributes are needed.

control_attribute_

Default:

allocation

control_attribute_allocation.initial_count = 2

 

control_attribute_allocation.max_count = -1 (unlimited)

 

control_attribute_allocation.incremental_count = -1 (number of attributes

 

will keep doubling on each allocation until it reaches max_count)

 

 

 

Forces an asynchronous send. When this parameter is set to 0, the TCP transport

 

will attempt to send data as soon as the internal send() function is called. When it is

 

set to 1, the transport will make a copy of the data to send and enqueue it in an

 

internal send buffer. Data will be sent as soon as the low-level socket buffer has

 

space.

 

Normally setting it to 1 delivers better throughput in a fast network, but will result

force_asynchronous_send

in a longer time to recover from various TCP error conditions. Setting it to 0 may

 

cause the low-level send() function to block until the data is physically delivered to

 

the lower socket buffer. For an application writing data at a very fast rate, it may

 

cause the caller thread to block if the send socket buffer is full. This could produce

 

lower throughput in those conditions (the caller thread could prepare the next

 

packet while waiting for the send socket buffer to become available).

 

Default: 0

 

 

 

The maximum size of a TCP segment.

 

This parameter is only supported on Linux architectures.

max_packet_size

By default, the maximum size of a TCP segment is based on the network MTU for

destinations on a local network, or on a default 576 for destinations on non-local

 

networks. This behavior can be changed by setting this parameter to a value

 

between 1 and 65535.

 

Default: -1 (default behavior)

 

 

7-11

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

Configures the sending of KEEP_ALIVE messages in TCP.

 

Setting this value to 1 causes a KEEP_ALIVE packet to be sent to the remote peer if a

 

long time passes with no other data sent or received.

 

This feature is implemented only on architectures that provide a low-level

 

implementation of the TCP keep-alive feature.

enable_keep_alive

On Windows systems, the TCP keep-alive feature can be globally enabled through

the system’s registry: \HKEY_LOCAL_MACHINE\SYSTEM\

 

 

CurrentControlSet\Tcpip\Parameters. Refer to MSDN documentation for more

 

details.

 

On Solaris systems, most of the TCP keep-alive parameters can be changed though

 

the kernel properties.

 

Default: 0

 

 

 

Specifies the interval of inactivity, in seconds, that causes TCP to generate a

keep_alive_time

KEEP_ALIVE message.

This parameter is only supported on Linux architectures.

 

 

Default: -1 (OS default value)

 

 

 

Specifies the interval, in seconds, between KEEP_ALIVE retries.

keep_alive_interval

This parameter is only supported on Linux architectures.

 

Default: -1 (OS default value)

 

 

 

The maximum number of KEEP_ALIVE retries before dropping the connection.

keep_alive_retry_count

This parameter is only supported on Linux architectures.

 

Default: -1 (OS default value)

 

 

 

Disables the TCP nagle algorithm.

disable_nagle

When this property is set to 1, TCP segments are always sent as soon as possible,

which may result in poor network utilization.

 

 

Default: 0

 

 

 

Bitmap that specifies the verbosity of log messages from the transport.

 

Logging values:

 

-1 (0xffffffff): do not change the current verbosity

 

0x00: silence

 

0x01: errors

 

0x02: warnings

 

0x04: local

 

0x08: remote

 

0x10: period

logging_verbosity_

0x80: other (used for control protocol tracing)

bitmap

Default: -1

 

 

Note: the logging verbosity is a global property shared across multiple instances of

 

the TCP transport. If you create a new TCP Transport instance with

 

logging_verbosity_bitmap different than -1, the change will affect all the other

 

instances as well.

 

The default TCP transport verbosity is errors and warnings.

 

Note: The option of 0x80 (other) is used only for tracing the internal control

 

protocol. Since the output is very verbose, this feature is enabled only in the debug

 

version of the TCP Transport library

 

(libnddstransporttcpd.so / LIBNDDSTRANSPORTD.LIB).

 

 

7-12

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

Maximum number of outstanding connection cookies allowed by the transport

 

when acting as server.

 

A connection cookie is a token provided by a server to a client; it is used to establish

 

a data connection. Until the data connection is established, the cookie cannot be

 

reused by the server.

outstanding_

To avoid wasting memory, it is good practice to set a cap on the maximum number

connection_cookies

of connection cookies (pending connections).

 

 

When the maximum value is reached, a client will not be able to connect to the

 

server until new cookies become available.

 

Range: 1 or higher, or -1 (which means an unlimited number).

 

Default: 100

 

 

 

Maximum lifespan (in seconds) of the cookies associated with pending connections.

outstanding_

If a client does not connect to the server before the lifespan of its cookie expires, it

will have to request a new cookie.

connection_cookies_

Range: 1 second or higher, or -1

life_span

 

 

Default : -1, which means an unlimited amount of time (effectively disabling the

 

feature).

 

 

 

A string that specifies the name of a file containing Certificate Authority certificates.

 

The file should be in PEM format. See the OpenSSL manual page for

tls.verify.ca_file

SSL_load_verify_locations for more information.

 

To enable TLS, ca_file or ca_path is required; both may be specified (at least one is

 

required).

 

 

 

A string that specifies paths to directories containing Certificate Authority

 

certificates. Files should be in PEM format and follow the OpenSSL-required

tls.verify.ca_path

naming conventions. See the OpenSSL manual page for

SSL_CTX_load_verify_locations for more information.

 

 

To enable TLS, ca_file or ca_path is required; both may be specified (at least one is

 

required).

 

 

tls.verify.verify_depth

Maximum certificate chain length for verification.

 

 

tls.verify.crl_file

Name of the file containing the Certificate Revocation List.

File should be in PEM format.

 

 

 

tls.cipher.cipher_list

List of available TLS ciphers. See the OpenSSL manual page for SSL_set_cipher_list

for more information on the format of this string.

 

 

 

 

List of available Diffie-Hellman (DH) key files.

 

For example: "foo.pem:512,bar.pem:256" means:

tls.cipher.

dh_param_files[0].file = foo.pem,

dh_param_files

dh_param_files[0].bits = 512,

 

dh_param_files[1].file = bar.pem,

 

dh_param_files[1].bits = 256

 

 

tls.cipher.engine_id

String ID of OpenSSL cipher engine to request.

 

 

 

A string containing an identifying certificate chain (in PEM format).

tls.identity.

An identifying certificate is required for secure communication.

The string must be sorted starting with the certificate to the highest level (root CA).

certificate_chain

 

 

Either certificate_chain or certificate_chain_file is required. You must set exactly one

 

of these. Do not set both of them (this would produce a configuration error.

 

 

7-13

Table 7.1 TCP/TLS Transport Properties (over LAN or WAN) — NDDS_Transport_TCPv4_Property_t

Property Name

 

(prefix with

Description

‘dds.transport.TCPv4.

 

tcp1.’)1

 

 

 

 

A string that specifies the name of a file containing an identifying certificate chain

 

(in PEM format). An identifying certificate is required for secure communication.

tls.identity.

The file must be sorted starting with the certificate to the highest level (root CA).

Optionally, a private key may be appended to this file. If a private key is not

certificate_chain_file

appended to this file, then either private_key or private_key_file is required.

 

 

Either certificate_chain or certificate_chain_file is required. You must set exactly

 

ONE of these. Do not set both of them (this would produce a configuration error.

 

 

tls.identity.

A string that specifies the password for private key.

private_key_password

 

 

 

 

A string containing a private key (in PEM format).

tls.identity.

Either private_key or private_key_file may be specified. Do not set both of them

private_key

(this would produce a configuration error). If both are unspecified (NULL), the

 

private key must be appended to the certificate chain file.

 

 

 

A string that specifies the name of a file containing a private key (in PEM format).

tls.identity.

Either private_key or private_key_file may be specified. Do not set both of them

private_key_file

(this would produce a configuration error). If both are unspecified (NULL), the

 

private key must be appended to the certificate chain file.

 

 

1. Assuming you used ‘dds.transport.TCPv4.tcp1’ as the alias to load the plugin. If not, change the prefix to match the string used with dds.transport.load_plugins.

2. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_UsersManual.pdf.

7-14

Chapter 8 Extending Routing Service with Adapters

As described in Section 2.4.8, adapters are pluggable components that allow Routing Service to consume and produce data for different data domains (e.g., Connext, JMS, Socket, etc.).

By default, Routing Service is distributed with a built-in Connext adapter. Any other adapter plugins must be provided as shared libraries or Java classes and registered within the

<adapter_library> tag.

The figure to the right describes the Routing Service adapter architecture.

Input adapters are used to collect data samples from different data domains, such as Connext or JMS. The input samples are processed by the Routing Service engine and are passed along to custom transformations if they are defined. Finally, the transformed data is provided to the output adapters.

The adapter plugin API is used to create new adapters; it is supported in C and Java.

The rest of this chapter describes:

Adapter Usage and Configuration (Section 8.1)

Adapter API And Entity Model (Section 8.2)

Creating New Adapters (Section 8.3)

8.1Adapter Usage and Configuration

Adapter plugins must be registered within an adapter library in the XML configuration file.

For example:

<?xml version="1.0"?> <dds>

<adapter_library name=”MyAdapterLibrary”>

8-1

<adapter_plugin name=”MyCAdapterPlugin”>

<dll>mycadapter</dll> <create_function>MyCAdapterPlugin_create</create_function>

</adapter_plugin>

<java_adapter_plugin name=”MyJavaAdapter”>

<class_name>com.rti.adapters.MyJavaAdapter</class_name> </java_adapter_plugin>

</adapter_library>

...

</dds>

C adapters are registered using the tag <adapter_plugin>; Java adapters use the tag

<java_adapter_plugin>.

Table 8.1 lists the tags allowed within <adapter_plugin>.

Table 8.2 lists the tags allowed within <java_adapter_plugin>.

Once the adapter plugins are registered, they can be used to create connections within a domain route (see Section 2.4.2).

For example:

<dds>

<routing_service name="Router1" group_name="Group1"> <domain_route name="DomainRoute1">

<connection_1 plugin_name=”MyAdapterLibrary::MyCAdapterPlugin”>

...

</connection_1>

<connection_2 plugin_name=”MyAdapterLibrary::MyJavaAdapter”>

...

</connection_2>

<session name="Session">

...

</session> </domain_route>

...

</routing_service> </dds>

Table 8.1 C Adapter Plugin Tags

Tags within

Description

Number

of Tags

<adapter_plugin>

 

Allowed

 

 

 

 

 

 

Required

 

 

Shared library containing the implementation of the adapter plugin.

 

 

The <dll> tag may specify the exact name of the file (for example,

 

 

lib/libmyadapter.so) or a general name (no file extension) which will be com-

 

 

pleted as follows:

 

<dll>

• <dll> value: dir/myadapter

1

• Final Path (UNIX-based systems): dir/libmyadapter.so

 

 

 

• Final Path (Windows systems): dir/myadapter.dll

 

 

If the library specified with the <dll> tag cannot be opened (because the

 

 

library path is not in the Path environment variable on a Windows system or

 

 

the LD_LIBRARY_PATH environment variables on a UNIX-based system),

 

 

Routing Service will look for the library in <Routing Service installation

 

 

directory>/bin/<architecture>.

 

 

 

 

8-2

Table 8.1 C Adapter Plugin Tags

 

Tags within

 

Description

Number

 

 

of Tags

 

<adapter_plugin>

 

 

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

Required

 

 

<create_function>

This tag must contain the name of the function used to create the adapter plu-

1

 

gin.

 

 

 

 

 

 

 

The function must be implemented in the adapter shared library.

 

 

 

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure the

 

 

 

parameters of the adapter. For example:

 

 

 

<property>

 

 

 

<value>

 

 

<property>

 

<element>

0 or 1

 

 

 

<name>username</name>

 

 

 

 

<value>myusername</value>

 

 

 

 

</element>

 

 

 

</value>

 

 

 

</property>

 

 

 

 

 

 

Table 8.2 Java Adapter Plugin Tags

 

 

 

 

 

 

 

Tags within

 

Description

Number

 

 

of Tags

 

<java_adapter_plugin>

 

 

Allowed

 

 

 

 

 

 

 

 

 

 

 

 

Required

 

 

 

 

Name of the class that implements the adapter plugin. For example:

 

 

 

 

com.rti.adapters.JMSAdapter

 

 

<class_name>

 

The classpath required to run the Java adapter must be part of the Rout-

1

 

 

 

ing Service JVM configuration. See Routing Service Tags (Table 2.2) for

 

 

 

 

additional information on JVM creation and configuration with the rout-

 

 

 

 

ing service.

 

 

 

 

 

 

 

 

 

Sequence of name/value(string) pairs that can be used to configure the

 

 

 

 

parameters of the adapter. For example:

 

 

 

 

<property>

 

 

 

 

<value>

 

 

<property>

 

<element>

0 or 1

 

 

 

<name>username</name>

 

 

 

 

<value>myusername</value>

 

 

 

 

</element>

 

 

 

 

</value>

 

 

 

 

</property>

 

 

 

 

 

 

8.2Adapter API And Entity Model

There are five main classes in the adapter class model:

1.Adapter: An Adapter is a factory for Connections. See Table 8.3, “Adapter Operations,” on page 8-4.

2.Connection: A Connection provides access to a data domain (such as a Connext domain or JMS provider network) and is a factory for Sessions, StreamReaders and StreamWriters.

In the built-in Connext adapter, a Connection is mapped to a DomainParticipant.

In an XML configuration file, connections are associated with the tags <connection_1> and <connection_2> within a domain route (see Section 2.4.2).

See Table 8.4, “Connection Operations,” on page 8-4.

8-3

3.Session: A Session is a concurrency unit within a connection that has an associated set of

StreamReaders and StreamWriters. Access to the StreamReaders and StreamWriters in the same Session is serialized by Routing Service (two StreamReaders/StreamWriters cannot be accessed concurrently).

In the built-in Connext adapter, a Session is mapped to a Publisher/Subscriber pair. In an XML file, Sessions are associated with the tag <session> (see Section 2.4.5). See Table 8.5, “Session Operations,” on page 8-7.

4.StreamReader: A StreamReader provides a way to read samples of a specific type from a data domain.

In the built-in Connext adapter, a StreamReader is mapped to a DataReader.

In an XML file, StreamReaders are associated with the tag <input> within <route> or <auto_route> (see Section 2.4.6).

See Table 8.6, “StreamReader Operations,” on page 8-7.

5.StreamWriter: A StreamWriter provides a way to write samples of a specific type in a data domain.

In the built-in Connext adapter, a StreamWriter is mapped to a DataWriter.

In an XML file, StreamWriters are associated with the tag <output> within <route> or <auto_route> (see Section 2.4.6).

See Table 8.7, “StreamWriter Operations,” on page 8-7.

Figure 8.1 describes the adapter class model.

Table 8.3 Adapter Operations

 

Operation

Description

 

 

 

 

 

Creates a new connection.

 

create_connection

Connection objects are created when the domain routes that contain them are enabled.

 

 

Implementation of this API is required.

 

 

 

 

 

Deletes a previously created connection.

 

delete_connection

Connection objects are deleted when the domain routes that contain them are disabled.

 

 

Implementation of this API is required.

 

 

 

 

 

Returns the Adapter’s version.

 

 

This method is only available in Java.

 

getVersion

In C, the version of the adapter is set on a member called plugin_version in the plugin

 

structure RTI_RoutingServiceAdapterPlugin (see Section 8.3.2).

 

 

 

 

The version of the adapter is only used for logging purposes.

 

 

Implementation of this API is required.

 

 

 

Table 8.4 Connection Operations

 

 

 

 

Operation

Description

 

 

 

 

 

 

 

connection_to_

Returns the string representation of a connection for logging purposes.

 

Implementation of this API is optional. If the API is not implemented, Routing Service

 

string

 

will use the fully qualified name of the adapter plugin.

 

 

 

 

 

 

 

Creates a new session.

 

create_session

Connection session objects are created when the associated routing service sessions are

 

enabled.

 

 

 

 

Implementation of this API is optional.

 

 

 

8-4

Figure 8.1 Adapter Class Model

Table 8.4 Connection Operations

Operation

Description

Deletes a previously created session.

delete_session

Connection session objects are deleted when the routing service sessions that contain them are disabled.

Implementation of this API is optional.

8-5

Table 8.4 Connection Operations

Operation

 

 

Description

 

 

 

 

 

 

 

 

 

 

 

Creates a new StreamReader within a routing service route.

 

 

 

This method is called when the route is enabled and the ‘creation mode’ condition asso-

 

ciated with the <input> tag becomes true (see Section 2.4.6.4).

 

 

create_

One of the parameters received by the create_stream_reader() operation is the Stream-

stream_reader

ReaderListener. The StreamReaderListener interface provides a callback which will be

 

used by the adapter to notify Routing Service of the existence of new data.

 

 

Implementation of this API is required only when there are routes using the adapter

 

to receive data.

 

 

 

 

 

 

 

 

 

Deletes a previously created StreamReader.

 

 

delete_

This method is called when the route is disabled or when the ‘creation mode’ condition

associated with the <input> tag becomes false (see Section 2.4.6.4).

 

stream_reader

 

Implementation of this API is required only when there are routes using the adapter

 

 

to receive data.

 

 

 

 

 

 

 

 

 

Creates a new StreamWriter within a routing service route.

 

 

create_

This method is called when the route is enabled and the ‘creation mode’ condition asso-

ciated with the <output> tag becomes true (see Section 2.4.6.4).

 

 

stream_writer

 

 

Implementation of this API is required only when there are routes using the adapter

 

 

to produce data.

 

 

 

 

 

 

 

 

 

Deletes a previously created StreamWriter.

 

 

delete_

This method is called when the route is disabled or when the ‘creation mode’ condition

associated with the <output> tag becomes false (see Section 2.4.6.4).

 

stream_writer

 

Implementation of this API is required only when there are routes using the adapter

 

 

to produce data.

 

 

 

 

 

 

 

Returns a StreamReader that is used by Routing Service to discover output streams. An

 

output stream is a stream to which StreamWriters can write data. Disposed scenarios,

get_output_

where an output stream disappears, are also notified using the discovery Stream-

Reader.

 

 

 

 

stream_discovery_

For additional information, see Stream Discovery (Section 8.2.2).

 

 

reader

 

 

Implementation of this API is optional. However, if none of the adapters in a domain

 

 

route implement the discovery API, the routes’ types must be declared in the configura-

 

tion file.

 

 

 

 

 

 

 

Returns a StreamReader that is used by Routing Service to discover input streams. An

 

input stream is a stream from which a StreamReader can read data. Disposed scenarios,

get_input_stream_

where an input stream disappears, are also notified using the discovery StreamReader.

For additional information, see Stream Discovery (Section 8.2.2).

 

 

discovery_reader

 

 

Implementation of this API is optional. However, if none of the adapters in a domain

 

 

route implement the discovery API, the routes’ types must be declared in the configura-

 

tion file.

 

 

 

 

 

 

 

 

Copies a type representation object (RoutingServiceTypeRepresentation).

 

 

The format of the type representation is given by the representation kind. For example,

 

if

the

representation

kind

is

copy_type_

RTI_ROUTING_SERVICE_TYPE_REPRESENTATION_DYNAMIC_TYPE,

the

type_representation will be a Connext TypeCode.

 

 

representation

 

 

This method is part of the adapter discovery API and is used by Routing Service to copy

 

 

the type representation of discovered streams (see Stream Discovery (Section 8.2.2).

 

 

Implementation

of this

API is optional and tied to the

implementation

of

 

get_input_stream_discovery_reader() and get_output_stream_discovery_reader().

 

 

 

 

 

 

 

8-6

Table 8.4 Connection Operations

Operation

Description

 

 

 

 

 

Deletes a previously created type-representation object.

delete_type_

This method is part of the adapter discovery API.

representation

Implementation of this API is optional and tied to the implementation of

 

 

get_input_stream_discovery_reader() and get_output_stream_discovery_reader().

 

 

 

Updates the connection’s configuration.

update

This method is called when the update command is received by the domain route con-

taining the connection (see Section 5.2.12).

 

 

Implementation of this API is optional.

 

 

Table 8.5 Session Operations

Operation

Description

Updates the configuration of a session.

update

This method is called when the update command is received by the routing service ses- sion (<session> tag) containing the adapter session (see Section 5.2.12).

Implementation of this API is optional.

Table 8.6 StreamReader Operations

 

 

Description

 

Operation

The StreamReader API is required only when the adapter is used to receive

 

data. Otherwise, it is optional.

 

 

 

 

 

Updates the configuration of a StreamReader providing a new set of properties.

 

update

This method is called after the update command is received by the routing service route

 

containing the StreamReader (see Section 5.2.12).

 

 

 

 

Implementation of this API is optional.

 

 

 

 

 

Reads a collection of data samples and sample infos from the StreamReader.

 

read

When Routing Service is done using the samples, it will 'return the loan' to the Stream-

 

Reader by calling return_loan().

 

 

 

 

Implementation of this API is required if the adapter is used to receive data.

 

 

 

 

 

Returns the loan on the read samples and infos.

 

return_loan

Routing Service calls this method to indicate that it is done accessing the collection of

 

data samples and sample infos obtained by an earlier invocation to read.

 

 

 

 

Implementation of this API is required if the adapter is used to receive data.

 

 

 

Table 8.7 StreamWriter Operations

 

 

 

 

 

Description

 

Operation

The StreamWriter API is only required when the adapter is used to pro-

 

duce data. Otherwise it is optional.

 

 

 

 

 

Updates the configuration of a StreamWriter providing a new set of properties.

 

update

This method is called after the update command is received by the routing service route

 

containing the StreamWriter (see Section 5.2.12).

 

 

 

 

Implementation of this API is optional.

 

 

 

 

 

Writes a collection of data samples and sample infos in the data domain associated with

 

write

the StreamWriter.

 

 

Implementation of this API is required if the adapter is used to produce data.

 

 

 

8-7

8.2.1Entity Creation

The sequence diagram in Figure 8.2 shows how the different Routing Service entities are created.

Figure 8.2 Entity Creation Sequence Diagram

:RoutingService

1: create()

:Adapter

 

<adapter_libraryname="adapters"> <adapter_plugin name="file">

<dll>fileadapter</dll> <create_function/>

</adapter_plugin> </adapter_library>

 

2: create_connection()

 

 

<domain_route>

3: create()

Connection1 :Connection

 

<connection_1

 

 

 

plugin_name="adapters::file">

 

 

 

<property>

4: create_connection()

 

 

<value>

5: create()

 

<element/>

 

Connection2 :Connection

</value>

 

 

 

</property>

 

 

 

</connection_1>

 

 

 

<connection_2

 

 

 

plugin_name="adapters::file">

 

 

 

<property>

 

 

 

<value>

 

 

 

<element/>

 

 

 

</value>

 

 

 

</property>

 

 

 

</connection_2>

 

 

 

</domain_route>

 

 

 

 

6: create_session()

7: create()

 

 

 

 

 

 

 

 

 

 

<session name="session">

 

 

 

:Session

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<value>

 

8: create_session()

 

 

 

 

 

 

 

 

 

 

<element/>

 

 

 

 

 

 

 

 

 

 

 

</value>

 

 

 

 

 

 

 

 

 

9: create()

 

 

 

</property>

 

 

 

 

 

 

 

 

 

 

:Session

 

 

 

 

 

 

 

 

 

 

 

</session>

10: create_stream_reader()

11: create()

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<route name="route">

 

 

:StreamReader

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<input connection="1">

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<value>

 

12: create_stream_writer()

 

 

 

 

 

 

 

 

 

 

 

<element/>

 

 

 

 

 

 

 

 

 

13: create()

:StreamWriter

 

 

 

</value>

 

 

 

 

 

 

 

 

 

 

 

 

 

</property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

</input>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<output>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<value>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<element/>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

</value>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

</property>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

</output>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

</route>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

An Adapter object is created when the first domain route that refers to it is enabled.

A Connection object is created when the domain route (<domain_route>) that contain it is enabled.

A Session object is created when the associated routing service session (<session>) is enabled.

A route’s StreamReader is created when the route is enabled and the ‘creation mode’ condition associated with the <input> tag becomes true (see Section 2.4.6.4).

A route’s StreamWriter is created when the route is enabled and the ‘creation mode’ condition associated with the <output> tag becomes true (see Section 2.4.6.4).

8-8

8.2.2Stream Discovery

A route cannot forward data until the type representations (e.g., TypeCode) associated with the input and output streams are available.

If a route refers to types that are not defined in the configuration file, Routing Service has to discover their type representation (e.g., TypeCode) before creating StreamReaders and StreamWriters. The adapter discovery API is used to provide stream and type information in a data domain to Routing Service.

The discovery API consists of four methods:

Connection::get_input_stream_discovery_reader()

Connection::get_output_stream_discovery_reader()

Connection::copy_type_representation()

Connection::delete_type_representation()

The first two methods provide access to StreamReaders used to discover streams in the data domain associated with a connection.

The input StreamReader (get_input_stream_discovery_reader()) provides information about input streams. An input stream is a stream from which a StreamReader read data. Disposed scenarios, where an input stream disappears, are also notified using the input StreamReader.

In the built-in Connext adapter, the input StreamReader is associated with the publication built- in DataReader of the DomainParticipant.

The output StreamReader (get_output_stream_discovery_reader()) provides information about output streams. An output stream is a stream to which StreamWriters can write data. Disposed scenarios, where an output stream disappears, are also notified using the output StreamReader.

In the built-in Connext adapter, the output StreamReader is associated with the subscription built-in DataReader of the DomainParticipant.

The

samples

provided

by

the

discovery

StreamReaders

have

the

type RoutingServiceStreamInfo.

 

 

 

 

 

 

struct RTI_RoutingServiceStreamInfo {
int disposed;

char * stream_name;

struct RTI_RoutingServiceTypeInfo type_info;

};

The dispose member is used to indicate whether the stream is a new discovered stream or a disposed stream.

The type_info member provides information about the type associated with the stream.

struct RTI_RoutingServiceTypeInfo {
char * type_name;

RTI_RoutingServiceTypeRepresentationKind type_representation_kind;

RTI_RoutingServiceTypeRepresentation type_representation;

};

The content associated with the type_representation depends on the type_representation_kind. For example, if the representation kind is

RTI_ROUTING_SERVICE_TYPE_REPRESENTATION_DYNAMIC_TYPE, the type_representation member will contain a Connext TypeCode. The method copy_type_representation() is used by Routing Service to copy the type representation associated with a discovered stream.

8-9

8.2.3Reading Data

Routing Service uses the session threads (there is one per <session> tag) to read data from StreamReaders.

Each session thread will block waiting for new data using a WaitSet. When a StreamReader receives new data, it will use the StreamReaderListener’s on_data_available() callback operation to wake up the session thread associated with it. After that, the session thread will invoke the StreamReader’s read() operation to get the new data.

The figure to the right describes how the session thread reads samples from a StreamReader.

8.3Creating New Adapters

Routing Service provides an adapter SDK in C and Java to support the creation of new adapter plugins.

The Routing Service Adapter SDK is distributed as a separate component that must be installed over an existing installation of Routing Service. For more information, see the Routing Service Adapter SDK SDK Installation Guide.

8.3.1Adapter SDK Components

After installing Routing Service Adapter SDK, the components in Table 8.8 will be available in the

Routing Service root folder.

Table 8.8 Adapter SDK Components

Component

Description

 

 

 

 

Release Notes and Installa-

Adapter SDK release notes and installation guide.

tion Guide

<Routing Service home>/doc/pdf

 

 

Adapter SDK Program-

Chapter 8 in the Routing Service User’s Manual (this chapter).

ming Guide

 

 

C and Java API specification in HTML and PDF format.

 

The C API specification describes the Adapter and Transformation API (see

API Specification

Chapter 4).

The Java API specification describes the Adapter API.

 

 

<Routing Service home>/ReadMe.html

 

<Routing Service home>/doc/pdf

 

 

8-10

Table 8.8 Adapter SDK Components

Component

 

Description

 

 

 

 

 

The SDK provides three buildable adapter implementations, two in C (file and

 

socket) and one in Java (JMS).

 

Adapter Sample Code

For instructions on compiling and using the sample adapters, see Section 4.9,

Section 4.10, and Section 4.11 in the Getting Started Guide.

 

 

Sample Code:

<Routing Service home>/adapters

 

Sample Configuration Files: <Routing Service home>/example/shapes

 

 

 

The SDK .jar file provides the necessary interfaces and support classes to imple-

SDK .jar file

ment Java adapters (see Section 8.3.5).

In addition, the JAR file also includes an implementation of a test adapter (Test-

(rtirsadapter.jar)

Adapter) that can be used to test new input adapters implementations.

 

 

JAR Location: <Routing Service home>/class/rtirsadapter.jar

 

 

SDK infrastructure shared

The infrastructure library provides environment (see Section 8.3.2.1) and proper-

ties management functions for C adapters.

library

The C adapters will have to link with this library.

([lib]rtirsinfrastruc-

Library Location: <Routing Service home>/bin/<architecture>/

ture[.dll,.so])

[lib]rtirsinfrastructure[.dll,.so]

 

 

 

 

The C adapters will have to include two SDK header files:

 

routingservice_adapter.h: This header file defines the adapter API.

 

routingservice_infrastructure.h: This header file defines the public interface of

SDK header files

the infrastructure library.

 

 

Header Location:

 

 

<Routing Service home>/include/routingservice_infrastructure.h

 

<Routing Service home>/include/routingservice_adapter.h

 

 

 

8.3.2C Adapter API

This section does not intend to give complete information on all the C API functions, but rather to describe the aspects of the API that are specific to the C language.

For detailed information about the C API, please see the online (HTML) Routing Service documentation.

Every adapter plugin will implement a plugin constructor (entry point to the shared library) that will be used by Routing Service to create a plugin instance.

typedef struct RTI_RoutingServiceAdapterPlugin * (

* RTI_RoutingServiceAdapterPlugin_CreateFcn)(

const struct RTI_RoutingServiceProperties * properties, RTI_RoutingServiceEnvironment * env);

The entry point function is specified in the configuration file using the tag <create_function> within <adapter_plugin> (see Section 8.1).

The structure RTI_RoutingServiceAdapterPlugin will contain the plugin implementation as a set of function pointers. This structure also encapsulates the plugin version information that will be used by Routing Service for logging purposes.

struct RTI_RoutingServiceAdapterPlugin {
int _init;

struct RTI_RoutingServiceVersion _rs_version;

/* The version of the adapter */

struct RTI_RoutingServiceVersion plugin_version;

RTI_RoutingServiceAdapterPlugin_DeleteFcn

adapter_plugin_delete;

8-11

/* Adapter API */
RTI_RoutingServiceAdapterPlugin_CreateConnectionFcn

adapter_plugin_create_connection; RTI_RoutingServiceAdapterPlugin_DeleteConnectionFcn

adapter_plugin_delete_connection;

/* Connection API */
RTI_RoutingServiceConnection_CreateSessionFcn

connection_create_session; RTI_RoutingServiceConnection_DeleteSessionFcn

connection_delete_session; RTI_RoutingServiceConnection_CreateStreamReaderFcn

connection_create_stream_reader; RTI_RoutingServiceConnection_DeleteStreamReaderFcn

connection_delete_stream_reader; RTI_RoutingServiceConnection_CreateStreamWriterFcn

connection_create_stream_writer; RTI_RoutingServiceConnection_DeleteStreamWriterFcn

connection_delete_stream_writer; RTI_RoutingServiceConnection_GetDiscoveryReaderFcn

connection_get_input_stream_discovery_reader; RTI_RoutingServiceConnection_GetDiscoveryReaderFcn

connection_get_output_stream_discovery_reader; RTI_RoutingServiceConnection_CopyTypeRepresentationFcn

connection_copy_type_representation; RTI_RoutingServiceConnection_DeleteTypeRepresentationFcn

connection_delete_type_representation; RTI_RoutingServiceConnection_GetAttributesFcn

connection_get_attributes; RTI_RoutingServiceConnection_ToStringFcn

connection_to_string; RTI_RoutingServiceAdapterEntity_UpdateFcn

connection_update;

/* Session API*/
RTI_RoutingServiceAdapterEntity_UpdateFcn

session_update;

/* Stream Reader API */
RTI_RoutingServiceStreamReader_ReadFcn

stream_reader_read; RTI_RoutingServiceStreamReader_ReturnLoanFcn

stream_reader_return_loan; RTI_RoutingServiceAdapterEntity_UpdateFcn

stream_reader_update;

/* Stream Writer API */
RTI_RoutingServiceStreamWriter_WriteFcn

stream_writer_write; RTI_RoutingServiceAdapterEntity_UpdateFcn

stream_writer_update;

void * user_object;

};

The adapter plugin instance created by the entry point function must be initialized with the macro RTI_RoutingServiceAdapterPlugin_initialize (part of the adapter API). For example:

struct RTI_RoutingServiceAdapterPlugin * MyAdapterPlugin_create(

const struct RTI_RoutingServiceProperties * properties, RTI_RoutingServiceEnvironment * env)

{

struct RTI_RoutingServiceAdapterPlugin * adapter = NULL;
struct RTI_RoutingServiceVersion version = {1,0,0,0};

adapter = calloc ( 1, sizeof (

struct RTI_RoutingServiceAdapterPlugin));

8-12

if (adapter == NULL) { RTI_RoutingServiceEnvironment_set_error (

env, "Memory allocation error"); return NULL;

}

RTI_RoutingServiceAdapterPlugin_initialize(adapter); adapter->plugin_version = version;

/* Assign the function pointers */

}

8.3.2.1Environment

The last parameter of each adapter API is the environment (RTI_RoutingServiceEnvironment). This parameter is used to get information about the Routing Service execution such as the version or the verbosity. The environment is also used by the adapter implementations to provide error notification.

8.3.2.2Adapter Verbosity

The C adapter implementations can access the verbosity level used to run Routing Service by using the following environment function:

RTI_RoutingServiceVerbosity RTI_RoutingServiceEnvironment_get_verbosity (

const RTI_RoutingServiceEnvironment * self);

 

The mapping

between the command-line option -verbosity and the

 

RTI_RoutingServiceVerbosity enumeration is as follows:

Table 8.9 Mapping between -verbosity and RTI_RoutingServiceVerbosity

 

 

 

 

-verbosity

RTI_RoutingServiceVerbosity

 

 

 

 

 

 

 

0

RTI_ROUTING_SERVICE_VERBOSITY_NONE

 

 

 

 

1

RTI_ROUTING_SERVICE_VERBOSITY_EXCEPTION

 

 

 

 

2

RTI_ROUTING_SERVICE_VERBOSITY_WARN

 

 

 

 

3 and 4

RTI_ROUTING_SERVICE_VERBOSITY_INFO

 

 

 

 

5 and 6

RTI_ROUTING_SERVICE_VERBOSITY_DEBUG

 

 

 

8.3.2.3Version Information

Routing Service and the different adapter implementations are identified by a version number.

The adapter version is provided to Routing Service using the member plugin_version in the RTI_RoutingServiceAdapterPlugin structure. This member must be initialized in the adapter entry point function; it is used by Routing Service for logging purposes.

The Routing Service version is provided to the C adapters through the environment. The adapters can access this information with the following function:

void RTI_RoutingServiceEnvironment_get_version( const RTI_RoutingServiceEnvironment * self, struct RTI_RoutingServiceVersion * version);

8.3.3My First C Adapter

This section shows how to create a simple C adapter on Windows and UNIX-based systems. It is not intended to give complete coverage of the entire adapter API, but rather to introduce the adapter technology and provide the basic process for developing a C adapter.

The new Adapter will be a simple file adapter where the input adapter reads lines from a text file and the output adapter saves the provided lines to an output text file.

8-13

A more flexible and complex file adapter that is able to work with structured information is provided under <Routing Service home>/adapters/file.

The source code and projects that you will create in the next sections are provided in <Routing

Service home>/adapters/tutorial/C.

8.3.3.1Setting the Environment on the Development Machine

There are a few things to take care of before you start developing the simple file adapter.

1.Install Connext; see the Routing Service Release Notes for the compatible version of Connext.

The C adapter will use TypeCode as the type representation format and DynamicData as the data representation format. This will require linking against the Connext libraries and including the header files defining the TypeCode and DynamicData APIs.

For information on how to install Connext, see the RTI Core Libraries and Utilities Getting Started Guide.1

As part of the installation process, make sure that NDDSHOME points to the Connext installation directory.

2.Set the environment variable ROUTINGSERVICEHOME

Set ROUTINGSERVICEHOME to the Routing Service installation directory. (Routing Service itself does not require that you set the environment variable. It is used to build and compile new adapters).

8.3.3.2Creating a Visual Studio Project (Only for Windows systems)

In this section you will create a Visual Studio project for the adapter dynamic library. We will use Microsoft® Visual Studio® 2008.

1.Start Microsoft Visual Studio 2008.

2.Select File, New, Project, Visual C++, Win32, Win32 Project. Name the project

SimpleFileAdapter and select a location.

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_GettingStarted.pdf.

8-14

3. Select Application Settings and choose DLL. Click Finish.

4.Create a new file called SimpleFileAdapter.c with the following content. This file will contain the adapter implementation.

/********* Simple File Adapter *****/

#include <stdio.h>
#include <string.h>
#ifdef RTI_WIN32

#include <process.h>
#else

#include <pthread.h>
#endif

#include "ndds/ndds_c.h"

#include "routingservice/routingservice_adapter.h"

#ifdef RTI_WIN32

/* Disable strtok, fopen warnings */
#pragma warning (disable : 4996 )
#define DllExport __declspec( dllexport )

#else

#define DllExport
#endif

/*--- Simple File Adapter: Adapter ---*/
/*
Entry point to the adapter plugin */

DllExport

struct RTI_RoutingServiceAdapterPlugin * SimpleFileAdapter_create(

const struct RTI_RoutingServiceProperties * properties, RTI_RoutingServiceEnvironment * env)

{

return NULL;

}

8-15

5. Add the new file to the project SimpleFileAdapter.

8-16

6.Right-click on SimpleFileAdapter, Properties.

In the Configuration combo box, select All Configurations.

Select Configuration Properties, C/C++, General.

Add the following to Additional Include Directories:

$(NDDSHOME)\include;$(NDDSHOME)\include\ndds; $(ROUTINGSERVICEHOME)\include

Select Configuration Properties, Linker, General; add the following to Additional Library Directories:

$(NDDSHOME)\lib\i86Win32VS2008; $(ROUTINGSERVICEHOME)\bin\i86Win32VS2008

Select Configuration Properties, Linker, Input; add the following to Additional Dependencies:

rtirsinfrastructure.lib nddsc.lib nddscore.lib netapi32.lib advapi32.lib user32.lib WS2_32.lib

In the Configuration combo box, select Debug.

Select Configuration Properties, C/C++, Preprocessor; replace the contents of Preprocessor Definitions with:

WIN32;WIN32_LEAN_AND_MEAN;NDDS_DLL_VARIABLE;RTI_WIN32;_DEBUG

In the Configuration combo box, select Release.

Select Configuration Properties, C/C++, Preprocessor; replace the contents of Preprocessor Definitions with:

WIN32;WIN32_LEAN_AND_MEAN;NDDS_DLL_VARIABLE;RTI_WIN32;NDEBUG

Click OK.

8-17

7. In the Solution Configuration combo box, select Release.

8.Build the SimpleFileAdapter project and verify that there are no errors.

8.3.3.3Creating an Adapter makefile [Only for UNIX-based systems)

In this section you will create a makefile to generate and compile the adapter shared library.

1.The makefile that you will generate is intended to be used with the GNU distribution of the make utility. On modern Linux systems, the make binary typically is GNU make. On other systems, GNU make is called gmake. The instructions below use gmake. Make sure that the GNU make binary is on your path before continuing.

2.Create a directory that will contain the adapter makefile and implementation. The rest of this section assumes that /opt/adapters/simplefile is the adapter directory.

3.In /opt/adapters/simplefile, create a file called makefile with the following content.

##########################################

# Makefile to build libsimplefileadapter.so

##########################################

ARCH = i86Linux2.6gcc4.1.1

c_cc = gcc c_ld = gcc

ifeq ($(DEBUG),1) c_cc_flags = -m32 -g else

c_cc_flags = -m32 endif

c_ld_flags = -m32 -static-libgcc syslibs = -ldl -lnsl -lm -lpthread

DEFINES_ARCH_SPECIFIC = -DRTI_UNIX -DRTI_LINUX

DEFINES = $(DEFINES_ARCH_SPECIFIC)

INCLUDES = -I. -I$(NDDSHOME)/include \
-I$(NDDSHOME)/include/ndds -I$(ROUTINGSERVICEHOME)/include

LIBS = \

-L$(NDDSHOME)/lib/$(ARCH) -L$(ROUTINGSERVICEHOME)/bin/$(ARCH) \

8-18

-lnddsc -lnddscore -lrtirsinfrastructure $(syslibs) $(extralibs)

COMMONSOURCES = SimpleFileAdapter.c

SHAREDLIB = lib/$(ARCH)/libsimplefileadapter.so

DIRECTORIES = lib.dir lib/$(ARCH).dir objs.dir objs/$(ARCH).dir

COMMONOBJS = $(COMMONSOURCES:%.c=objs/$(ARCH)/%.o)

$(ARCH) : $(DIRECTORIES) $(COMMONOBJS) $(SHAREDLIB)

$(SHAREDLIB) : $(COMMONOBJS)

$(c_cc) $(c_ld_flags) -shared -o $@ $^ $(LIBS)

objs/$(ARCH)/%.o : %.c

$(c_cc) $(c_cc_flags) -o $@ $(DEFINES) $(INCLUDES) -c $<

# Here is how we create those subdirectories automatically. %.dir :

@echo "Checking directory $*" @if [ ! -d $* ]; then \

echo "Making directory $*"; \ mkdir -p $* ; \

fi;

clean:

@rm -rf ./objs @rm -rf ./lib

The above makefile assumes that the architecture is i86Linux2.6gcc4.1.1. If you are build- ing for a different architecture, you can use the above makefile as an example.

4.Create a new file called SimpleFileAdapter.c with the following content. This file will contain the adapter implementation.

#include <stdio.h> #include <string.h>
#ifdef RTI_WIN32 #include <process.h>
#else

#include <pthread.h>
#endif

#include "ndds/ndds_c.h"

#include "routingservice/routingservice_adapter.h"

#ifdef RTI_WIN32

/* Disable strtok, fopen warnings */
#pragma warning( disable : 4996 )
#define DllExport __declspec( dllexport )
#else

#define DllExport
#endif

/* Entry point to the adapter plugin */
DllExport

struct RTI_RoutingServiceAdapterPlugin * SimpleFileAdapter_create(

const struct RTI_RoutingServiceProperties * properties, RTI_RoutingServiceEnvironment * env)

{

return NULL;

}

5.Compile the SimpleFileAdapter skeleton by executing gmake from the adapter directory.

>gmake

8-19

After compilation, you will find the adapter library in /opt/adapters/simplefile/lib/ <architecture>. The next few sections will show you how to complete the adapter implementation.

8.3.3.4Initializing the Adapter Entry Point Function

Every adapter plugin must implement a plugin constructor (entry point to the dynamic library) that will be used by Routing Service to create a plugin instance (see Section 8.3.2). In this example, the entry point is the function SimpleFileAdapter_create in the file SimpleFileAdapter.c. You have to initialize this function to create a new plugin.

/* Plugin destructor */

void SimpleFileAdapter_delete(

struct RTI_RoutingServiceAdapterPlugin * adapter, RTI_RoutingServiceEnvironment * env)

{

free(adapter);

}

/* Entry point to the adapter plugin */

DllExport struct RTI_RoutingServiceAdapterPlugin * SimpleFileAdapter_create(

const struct RTI_RoutingServiceProperties * properties, RTI_RoutingServiceEnvironment * env)

{

struct RTI_RoutingServiceAdapterPlugin * adapter = NULL;
struct RTI_RoutingServiceVersion version = {1,0,0,0};

int verbosity;

verbosity = RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) { printf("CALL SimpleFileAdapter_create\n");

}

adapter = calloc ( 1, sizeof (

struct RTI_RoutingServiceAdapterPlugin));
if (adapter == NULL) {

RTI_RoutingServiceEnvironment_set_error ( env, "Memory allocation error");

return NULL;

}

RTI_RoutingServiceAdapterPlugin_initialize(adapter); adapter->plugin_version = version;

/* Assign the function pointers */ adapter->adapter_plugin_delete = SimpleFileAdapter_delete;

return (struct RTI_RoutingServiceAdapterPlugin *) adapter;

}

The structure RTI_RoutingServiceAdapterPlugin contains the plugin implementation as a set of function pointers. For now, you only need to implement adapter_plugin_delete that deletes the plugin instances created by SimpleFileAdapter_create(). You will initialize the other pointers in the plugin structure as you implement the adapter functionality.

The entry point function receives two parameters: the adapter properties and the environment, env.

The properties parameter (not used by the SimpleFileAdapter) is used to configure the adapter instance. The values contained in this parameter are provided as (name,value) pairs using the tag <property> within <adapter_plugin> (see Adapter Usage and Configuration (Section 8.1)).

8-20

The environment parameter, env, is part of every function in the adapter API. This parameter is used to get information about the Routing Service execution such as the version or the verbosity. In addition, the environment is also used to notify Routing Service of any error in the adapter execution.

8.3.3.5Implementing the Adapter Connection

The adapter plugin instances are connection factories. Connection objects provide access to data domains such as Connext domains or JMS network providers and they are configured using the XML tags <connection_1> and <connection_2> in a <domain_route> (see Section 2.4.2). In the SimpleFileAdapter example, the connection objects will provide access to a directory on your computer’s file system.

The next step consist on implementing the functions that create and delete a connection. Insert the following code in the “Simple File Adapter: Connection“ section of SimpleFileAdapter.c.

/* Connection */

struct SimpleFileAdapterConnection { char * directory;

};

/* Deletes a connection */

void SimpleFileAdapter_delete_connection(

struct RTI_RoutingServiceAdapterPlugin * adapter, RTI_RoutingServiceConnection connection, RTI_RoutingServiceEnvironment * env)

{

struct SimpleFileAdapterConnection * cx =

(struct SimpleFileAdapterConnection *) connection;

int verbosity; verbosity =

RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {

printf("CALL SimpleFileAdapter_delete_connection\n");

}

if (cx->directory != NULL) {
free(cx->directory);

}

free(cx);

}

/* Creates a connection */ RTI_RoutingServiceConnection SimpleFileAdapter_create_connection(

struct RTI_RoutingServiceAdapterPlugin * adapter, const char * routing_service_name,

const char * routing_service_group_name,

const struct RTI_RoutingServiceStreamReaderListener * input_disc_listener,

const struct RTI_RoutingServiceStreamReaderListener * output_disc_listener,

const struct RTI_RoutingServiceTypeInfo

**registeredTypes, int registeredTypeCount,

const struct RTI_RoutingServiceProperties

*properties,

RTI_RoutingServiceEnvironment * env)

{

const char * directory;

struct SimpleFileAdapterConnection * cx; int verbosity;

verbosity =

8-21

RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {

printf("CALL SimpleFileAdapter_create_connection\n");

}

cx = calloc ( 1,

sizeof ( struct SimpleFileAdapterConnection));
if (cx == NULL) {

RTI_RoutingServiceEnvironment_set_error( env, "Memory allocation error");

return NULL;

}

directory = RTI_RoutingServiceProperties_lookup_property( properties,"directory");

if (directory == NULL) {
RTI_RoutingServiceEnvironment_set_error(

env, "directory property is required"); free(cx);

return NULL;

}

cx->directory = (char *)malloc(strlen(directory)+1);
if (cx->directory == NULL) {

RTI_RoutingServiceEnvironment_set_error ( env, "Memory allocation error");

free(cx);
return NULL;

}

strcpy(cx->directory, directory);
return cx;

}

From the implementation, you can see that the connection object encapsulates the name of the directory from which the StreamReaders and StreamWriters will read and write files.

The value of the RTI_RoutingServiceAdapterPlugin structure created in SimpleFileAdapter_create() must be updated to contain the two new functions.

adapter->adapter_plugin_create_connection = SimpleFileAdapter_create_connection; adapter->adapter_plugin_delete_connection = SimpleFileAdapter_delete_connection;

8.3.3.6Implementing the StreamReader

The connection objects are factories of StreamReaders. A StreamReader provides a way to read data samples of a specific type from a data domain.

In the configuration file, StreamReaders are associated with the tag <input> within <route> or <auto_route> (see Section 2.4.6).

The StreamReaders created by the SimpleFileAdapter connections read text files from the connection directory.

The data samples provided to Routing Service (using the read operation) are DynamicData with the following IDL type:

struct TextLine { string<1024> value;

};

When a SimpleFileAdapter StreamReader is created, the name of the file is the input stream name with a .txt extension. You can use the read_period property to control how often the StreamReader notifies Routing Service about new lines. For example:

<route name="route"> <input connection="1">

<stream_name> HelloWorld </stream_name>

8-22

<registered_type_name> TextLine

</registered_type_name> <property>

<value>

<element> <name>read_period</name> <value>1000</value>

</element>

</value>

</property>

</input>

...

</route>

In the above example, the input StreamReader will read lines from a file called HelloWorld.txt and provide one line per second to Routing Service.

The next step is to implement the StreamReader functionality. You will implement five new functions:

SimpleFileAdapterStreamReader_read(): This function will be called by Routing Service after being notified that the are new lines available. Although the signature of the function allows returning more than one sample (line), for the sake of simplicity, the implementation only returns one line each time the function is called.

SimpleFileAdapterStreamReader_return_loan: The loan on the samples provided by

SimpleFileAdapterStreamReader_read() is returned to the adapter using this function. The SimpleFileAdapter implementation of return_loan() is empty because:

The read operation does not create new samples and always returns a single sample stored in the StreamReader.

Two calls to SimpleFileAdapterStreamReader_read() cannot occur in parallel.

SimpleFileAdapterStreamReader_run: Routing Service will not call the read operation until it is notified of the presence of new data (see Section 8.2.3). To provide data notification, the StreamReader implementation creates a thread that wakes up after read_period and notifies Routing Service of new data if the end of the file has not been reached yet. SimpleFileAdapterStreamReader_run is the function executed by the notification thread.

SimpleFileAdapterConnection_delete_stream_reader: This function is called to destroy a StreamReader. The implementation will finalize the notification thread and close the file handle.

SimpleFileAdapterConnection_create_stream_reader: This function is called when a new StreamReader is created. Among other things, the implementation will open the file that will be read and create the notification thread.

Insert the following code in the “Simple File Adapter: StreamReader“ section of

SimpleFileAdapter.c.

/* StreamReader */

struct SimpleFileAdapterStreamReader {
int run;

#ifdef RTI_WIN32 HANDLE thread;
#else

pthread_t thread;
#endif

DDS_DynamicData * sample[1];
struct DDS_Duration_t readPeriod;

struct RTI_RoutingServiceStreamReaderListener listener;

FILE * fHandle;

};

8-23

/* Returns sample loan */

void SimpleFileAdapterStreamReader_return_loan( RTI_RoutingServiceStreamReader stream_reader, RTI_RoutingServiceSample * sample_list, RTI_RoutingServiceSampleInfo * info_list,

int count, RTI_RoutingServiceEnvironment * env)

{

int verbosity;
verbosity =

RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {

printf(

"CALL SimpleFileAdapterStreamReader_return_loan\n");

}

return;

}

/* Reads one line from the StreamReader file */ void SimpleFileAdapterStreamReader_read(

RTI_RoutingServiceStreamReader stream_reader, RTI_RoutingServiceSample ** sample_list, RTI_RoutingServiceSampleInfo ** info_list,

int * count, RTI_RoutingServiceEnvironment * env)

{

DDS_ReturnCode_t retCode;
char line[2048];

char * str;

struct SimpleFileAdapterStreamReader * self = (struct SimpleFileAdapterStreamReader *)

stream_reader; int verbosity;
int length;

verbosity = RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {

printf("CALL SimpleFileAdapterStreamReader_read\n");

}

*count = 0;
*sample_list = NULL;

/* We don't provide sample info in this adapter, which is an optional feature

*/

*info_list = NULL;

DDS_DynamicData_clear_all_members(self->sample[0]);

str = fgets(line, sizeof(line), self->fHandle);

if (!str) { return;

}

length = strlen(str);

if (length > 0 && str[length-1] == '\n') { str[length-1] = '\0';

if (length > 1 && str[length-2] == '\r') { str[length-2] = '\0';

}

8-24

}

retCode = DDS_DynamicData_set_string( self->sample[0], "value",

DDS_DYNAMIC_DATA_MEMBER_ID_UNSPECIFIED, line);

if (retCode != DDS_RETCODE_OK) {
RTI_RoutingServiceEnvironment_set_error(

env, "Error assigning value=%s", line);
return;

}

*sample_list = (RTI_RoutingServiceSample *)self->sample; *count = 1;

return;

}

/* Notification thread

This thread will notify of data availability in the file. */

void * SimpleFileAdapterStreamReader_run( void * threadParam)

{

struct SimpleFileAdapterStreamReader * self =

(struct SimpleFileAdapterStreamReader *) threadParam;

while (self->run) {
NDDS_Utility_sleep(&self->readPeriod);

if (!feof(self->fHandle)) {
self->listener.on_data_available( self, self->listener.listener_data);

}

}

return NULL;

}

/* Deletes a StreamReader */

void SimpleFileAdapterConnection_delete_stream_reader( RTI_RoutingServiceConnection connection, RTI_RoutingServiceStreamReader stream_reader, RTI_RoutingServiceEnvironment * env)

{

struct SimpleFileAdapterStreamReader * reader =

(struct SimpleFileAdapterStreamReader *) stream_reader;
#ifndef RTI_WIN32

void * value = NULL;
#endif

int verbosity;

verbosity = RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {
printf(

"CALL SimpleFileAdapterConnection_delete_stream_reader\n");

}

reader->run = 0;

#ifdef RTI_WIN32 WaitForSingleObject(reader->thread, INFINITE);

#else

pthread_join(reader->thread, &value);
#endif

if (reader->fHandle) { fclose(reader->fHandle);

8-25

}

if (reader->sample[0]) { DDS_DynamicData_delete(reader->sample[0]);

}

free(reader);

}

/* Creates a StreamReader */ RTI_RoutingServiceStreamReader SimpleFileAdapterConnection_create_stream_reader(

RTI_RoutingServiceConnection connection, RTI_RoutingServiceSession session,

const struct RTI_RoutingServiceStreamInfo * stream_info, const struct RTI_RoutingServiceProperties * properties,

const struct RTI_RoutingServiceStreamReaderListener * listener, RTI_RoutingServiceEnvironment * env)

{

const char * readPeriodStr;
unsigned int readPeriod;
char * file;

struct SimpleFileAdapterConnection * self =

(struct SimpleFileAdapterConnection *)connection;
struct SimpleFileAdapterStreamReader * reader = NULL;
struct DDS_DynamicDataProperty_t dynamicDataProps =

DDS_DynamicDataProperty_t_INITIALIZER;
int error = 0;

#ifndef RTI_WIN32 pthread_attr_t threadAttr;

#endif

int verbosity; verbosity =

RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {
printf(

"CALL SimpleFileAdapterConnection_create_stream_reader\n");

}

/* Create StreamReader */ reader = calloc(1, sizeof(

struct SimpleFileAdapterStreamReader));

if (reader == NULL) {
RTI_RoutingServiceEnvironment_set_error(

env, "Memory allocation error");
return NULL;

}

reader->sample[0] = DDS_DynamicData_new(
(struct DDS_TypeCode *) stream_info->type_info.type_representation, &dynamicDataProps);

if (reader->sample[0] == NULL) {
RTI_RoutingServiceEnvironment_set_error(

env, "Failure creating dynamic data sample");
free(reader);

return NULL;

}

/* Open input file */

file = malloc(strlen(self->directory) +
strlen("/") + strlen(stream_info->stream_name) + strlen(".txt") + 1);

8-26

if (file == NULL) {
RTI_RoutingServiceEnvironment_set_error(

env, "Memory allocation error");
DDS_DynamicData_delete(reader->sample[0]);
free(reader);

return NULL;

}

sprintf(file, "%s/%s.txt",

self->directory, stream_info->stream_name);

reader->fHandle = fopen(file, "r");

if (reader->fHandle == NULL) {
RTI_RoutingServiceEnvironment_set_error(

env, "Error opening %s", file); free(file);

DDS_DynamicData_delete(reader->sample[0]);
free(reader);

return NULL;

}

free(file);

/* Creates notification thread */ readPeriodStr =

RTI_RoutingServiceProperties_lookup_property( properties, "read_period");

if (readPeriodStr != NULL) {
readPeriod = atoi(readPeriodStr);

} else {

readPeriod = 1000; /* 1 Sec */

}

reader->readPeriod.sec = readPeriod / 1000;
reader->readPeriod.nanosec =

(readPeriod % 1000) * 1000000;
reader->run = 1;
reader->listener = *listener;

#ifdef RTI_WIN32

reader->thread =
(HANDLE) _beginthread( (void(__cdecl*)(void*)) SimpleFileAdapterStreamReader_run, 0, (void*)reader);

if (!reader->thread) { error = 1;

}

#else pthread_attr_init(&threadAttr); pthread_attr_setdetachstate(

&threadAttr, PTHREAD_CREATE_JOINABLE);

error = pthread_create( &reader->thread, &threadAttr,

SimpleFileAdapterStreamReader_run, (void *)reader);

pthread_attr_destroy(&threadAttr);
#endif

if (error) {
RTI_RoutingServiceEnvironment_set_error(

env, "Error creating notification thread");
DDS_DynamicData_delete(reader->sample[0]);

8-27

free(reader);
fclose(reader->fHandle);
return NULL;

}

return reader;

}

The value of the RTI_RoutingServiceAdapterPlugin structure created in SimpleFileAdapter_create() must be updated to contain the StreamReader functions.

adapter->connection_create_stream_reader = SimpleFileAdapterConnection_create_stream_reader;

adapter->connection_delete_stream_reader = SimpleFileAdapterConnection_delete_stream_reader;

adapter->stream_reader_read = SimpleFileAdapterStreamReader_read;

adapter->stream_reader_return_loan = SimpleFileAdapterStreamReader_return_loan;

8.3.3.7Implementing the StreamWriter

The connection objects are factories of StreamWriters. A StreamWriter provides a way to write samples of a specific type into a data domain.

In the configuration file, StreamWriters are associated with the tag <output> within <route> or <auto_route> (see Section 2.4.6).

The SimpleFileAdapter StreamWriters create new files into the connection directory and store the lines read from the routes’ inputs.

The data samples provided to the write operation of the StreamWriters are DynamicData with the following IDL type:

struct TextLine { string<1024> value;

};

When a SimpleFileAdapter StreamWriter is created, the name of the file is the output stream name with “.txt” extension. For debugging purposes, the StreamWriter can be configured to print the written samples on the console:

<route name="route">

...

<output> <stream_name>HelloWorld</stream_name> <registered_type_name>

TextLine </registered_type_name> <property>

<value>

<element> <name>print_to_stdout</name> <value>1</value>

</element>

</value>

</property>

</output>

</route>

In the above example, the output StreamWriter will store the lines provided by Routing Service on a file called HelloWorld.txt. It will also print the lines on the screen.

Insert the following code in the “Simple File Adapter: StreamWriter“ section of

SimpleFileAdapter.c.

/* StreamWriter */

8-28

struct SimpleFileAdapterStreamWriter {
int printToStdout;

FILE * fHandle;

};

int SimpleFileAdapterStreamWriter_write( RTI_RoutingServiceStreamWriter stream_writer, const RTI_RoutingServiceSample * sample_list, const RTI_RoutingServiceSampleInfo * info_list, int count,

RTI_RoutingServiceEnvironment * env)

{

int i, samplesWritten;
DDS_DynamicData * sample;
DDS_ReturnCode_t retCode;
char * line;

struct SimpleFileAdapterStreamWriter * self =
(struct SimpleFileAdapterStreamWriter *) stream_writer;

int verbosity;
verbosity =

RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) {
printf("CALL SimpleFileAdapterStreamWriter_write\n");

}

if (!self->fHandle) {
return 0;

}

samplesWritten = 0;

for (i=0; i<count; i++) {

sample = (DDS_DynamicData *)sample_list[i];
line = NULL;

retCode = DDS_DynamicData_get_string( sample, &line, NULL, "value",

DDS_DYNAMIC_DATA_MEMBER_ID_UNSPECIFIED);

if (retCode != DDS_RETCODE_OK) {
RTI_RoutingServiceEnvironment_set_error(

env, "Error assigning value");

} else { samplesWritten++;

}

fputs(line, self->fHandle);
fputs("\n”, self->fHandle);
fflush(self->fHandle);

if (self->printToStdout) {
printf("%s\n",line);
fflush(stdout);

}

DDS_String_free(line);

}

return samplesWritten;

}

/*

* Deletes a StreamWriter */

void SimpleFileAdapterConnection_delete_stream_writer(

8-29

RTI_RoutingServiceConnection connection, RTI_RoutingServiceStreamWriter stream_writer, RTI_RoutingServiceEnvironment * env)

{

struct SimpleFileAdapterStreamWriter * writer = (struct SimpleFileAdapterStreamWriter *)

stream_writer;

int verbosity; verbosity =

RTI_RoutingServiceEnvironment_get_verbosity(env); if (verbosity ==

RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) { printf(

"CALL SimpleFileAdapterConnection_delete_stream_writer\n");

}

if (writer->fHandle) { fclose(writer->fHandle);

}

free(writer);

}

/*

* Creates a StreamWriter */

RTI_RoutingServiceStreamWriter SimpleFileAdapterConnection_create_stream_writer(

RTI_RoutingServiceConnection connection, RTI_RoutingServiceSession session,

const struct RTI_RoutingServiceStreamInfo * stream_info,

const struct RTI_RoutingServiceProperties * properties,

RTI_RoutingServiceEnvironment * env)

{

const char * printToStdoutStr; char * file;

struct SimpleFileAdapterConnection * self =

(struct SimpleFileAdapterConnection *)connection; struct SimpleFileAdapterStreamWriter * writer = NULL; int verbosity;

verbosity = RTI_RoutingServiceEnvironment_get_verbosity(env);

if (verbosity == RTI_ROUTING_SERVICE_VERBOSITY_DEBUG) { printf(

"CALL SimpleFileAdapterConnection_create_stream_writer\n");

}

/* Create StreamWriter */ writer = calloc(1,

sizeof(struct SimpleFileAdapterStreamWriter));

if (writer == NULL) { RTI_RoutingServiceEnvironment_set_error(

env, "Memory allocation error"); return NULL;

}

/* Open output file */

file = malloc(strlen(self->directory) + strlen("/") + strlen(stream_info->stream_name) +

strlen(".txt") + 1);

8-30

if (file == NULL) { RTI_RoutingServiceEnvironment_set_error(

env, "Memory allocation error"); free(writer);

return NULL;

}

sprintf(file,"%s/%s.txt",

self->directory, stream_info->stream_name);

writer->fHandle = fopen(file, "w+");

if (writer->fHandle == NULL) { RTI_RoutingServiceEnvironment_set_error(

env, "Error opening %s", file); free(file);

free(writer); return NULL;

}

free(file);

/* Creates notification thread */ printToStdoutStr =

RTI_RoutingServiceProperties_lookup_property( properties, "print_to_stdout");

if (printToStdoutStr != NULL) { writer->printToStdout = atoi(printToStdoutStr);

} else { writer->printToStdout = 0;

}

return writer;

}

The value of the RTI_RoutingServiceAdapterPlugin structure created in SimpleFileAdapter_create() must be updated to contain the StreamWriter functions.

adapter->connection_create_stream_writer = SimpleFileAdapterConnection_create_stream_writer;

adapter->connection_delete_stream_writer = SimpleFileAdapterConnection_delete_stream_writer;

adapter->stream_writer_write = SimpleFileAdapterStreamWriter_write;

8.3.3.8Running the SimpleFileAdapter

This section describes the steps required to use and run the SimpleFileAdapter with Routing Service. You will create a configuration file with a single route that reads a HelloWorld text file from an input directory and saves it into an output directory.

1.If you have not done it yet, compile and build the SimpleFileAdapter.

2.Under the adapter project directory (c:\adapters\SimpleFileAdapter1 on Windows systems; /opt/adapters/simplefile1 on UNIX-based systems) create two directories called input and output.

3.In the input directory create a file called HelloWorld.txt with the following content.

Hello World 1!

Hello World 2!

Hello World 3!

Hello World 4!

Hello World 5!

Hello World 6!

Hello World 7!

Hello World 8!

1. Your directory may be different if you did not use the default locations.

8-31

Hello World 9!

Hello World 10!

4.In the adapter project directory create a Routing Service XML configuration file called simple_file_adapter.xml with the following content.

Replace the value of the “directory” property under both connections with the location of the input and output directories.

Replace the content of the dll tag under adapter_plugin with the location of the release version of the SimpleFileAdapter shared library.

<?xml version="1.0"?> <dds>

<adapter_library name="adapters"> <adapter_plugin name="simple_file">

<dll>

c:\adapters\SimpleFileAdapter\Release\SimpleFileAdapter.dll

</dll> <create_function>

SimpleFileAdapter_create </create_function> </adapter_plugin>

</adapter_library>

<types>

<struct name="TextLine"> <member name=

"value" type="string" stringMaxLength="2048"/> </struct>

</types>

<routing_service name="file_to_file"> <domain_route name="domain_route">

<connection_1 plugin_name="adapters::simple_file"> <registered_type name=

"TextLine" type_name="TextLine"/> <property>

<value>

<element>

<name>directory</name>

<value>

c:\adapters\SimpleFileAdapter\input

</value>

</element>

</value>

</property> </connection_1>

<connection_2 plugin_name="adapters::simple_file"> <registered_type name=

"TextLine" type_name="TextLine"/> <property>

<value>

<element>

<name>directory</name>

<value>

c:\adapters\SimpleFileAdapter\output

</value>

</element>

</value>

</property> </connection_2>

<session name="session"> <route name="route">

8-32

<input connection="1"> <stream_name> HelloWorld </stream_name>

<registered_type_name> TextLine

</registered_type_name> </input>

<output> <stream_name>

HelloWorld </stream_name> <registered_type_name>

TextLine </registered_type_name> <property>

<value>

<element>

<name> print_to_stdout

</name>

<value>1</value>

</element>

</value>

</property>

</output>

</route>

</session> </domain_route>

</routing_service> </dds>

5.Start Routing Service by entering the following in a command shell.

On UNIX-based systems:

>cd <SimpleFileAdapter project directory>

>$ROUTINGSERVICEHOME/scripts/rtiroutingservice

-cfgFile simple_file_adapter.xml -cfgName file_to_file

On Windows systems:

>cd <SimpleFileAdapter project directory>

>%ROUTINGSERVICEHOME%\scripts\rtiroutingservice -cfgFile simple_file_adapter.xml -cfgName file_to_file

6.On the screen you will see:

RTI Routing Service <version> started (with name file_to_file)

Hello World 1!

Hello World 2!

Hello World 3!

Hello World 4!

Hello World 5!

Hello World 6!

Hello World 7!

Hello World 8!

Hello World 9!

Hello World 10!

7.Verify that a file called HelloWorld.txt has been generated into the output directory. The content of this file should de identical to the content of the same file in the input directory.

8.3.4Debugging C Adapters

When you develop a custom adapter you will need to debug it and test it. This section talks about the tools and APIs that you have available to debug and detect problems in Routing Service adapters written in C.

8-33

The first debugging capability is provided by the Routing Service SDK. The adapter SDK provides a way to access the verbosity level of Routing Service through the usage of the environment function RTI_RoutingServiceEnvironment_get_verbosity. It is highly recommendable that as part of the adapter implementation you instrument the code by adding status messages that will be printed with the INFO and DEBUG verbosity levels. This level of instrumentation will help you to capture run-time information for troubleshooting.

The second debugging capability is provided by third party tools. On a Windows system, you can debug the adapter shared libraries using Visual Studio. On a UNIX-based system, you can use GDB, the GNU Project debugger.

8.3.4.1Debugging the Adapter with Visual Studio—only for Windows Systems

Let’s see how to debug the adapter library with Visual Studio 2008 using the SimpleFileAdapter implemented in Section 8.3.3.

1.Start Microsoft Visual Studio 2008 and open the solution SimpleFileAdapter.

2.In the Solution Configuration combo box select Debug configuration and recompile the SimpleFileAdapter project.

3.Edit simple_file_adapter.xml, the configuration file generated in Section 8.3.3.8. Replace the library in the <dll> tag with the debug version of the adapter. For example:

<dll>c:\adapters\SimpleFileAdapter\Debug\SimpleFileAdapter.dll</dll>

4. Right-click on SimpleFileAdapter, Properties

In the configuration combo box select Debug.

Under Configuration Properties, Debugging; go to “Command” and add the following:

$(ROUTINGSERVICEHOME)\bin\i86Win32VS2008\rtiroutingservice

Under Configuration Properties, Debugging; go to “Command Arguments” and add the following:

-cfgFile c:\adapters\SimpleFileAdapter\simple_file_adapter.xml1 -cfgName file_to_file

1. The location of your configuration file may be different. Replace the value with the right location.

8-34

Click OK.

5.Open the file SimpleFileAdapter.c and insert breakpoints in the functions that you want to debug. Then press F5 to run Routing Service and debug the adapter.

If you get an information window that says there is no debugging information in rtiroutingservice, press YES. Although rtiroutingservice does not have debugging symbols, your adapter was built with debug information and you should not have any problems debugging it.

8.3.4.2Debugging the Adapter with GDB—only for UNIX-based systems

Let’s see how to debug the adapter library with gdb using the SimpleFileAdapter implemented in Section 8.3.3.

1.Go to the directory containing the SimpleFileAdapter makefile and build the debug ver- sion of the shared library as follows:

>gmake clean

>gmake DEBUG=1

The debug version of the adapter replaces the release version because is generated in the same location.

2.Edit the configuration file simple_file_adapter.xml generated in Section 8.3.3.8 and replace the library in the <dll> tag with the debug version of the adapter. For example:

<dll>/opt/adapters/simplefile/lib/i86Linux2.6gcc4.1.1/libsimplefileadapter.so</dll>

3.Run gdb:

>gdb $ROUTINGSERVICEHOME/bin/i86Linux2.6gcc4.1.1/rtiroutingservice

4.Insert breakpoints in the functions that you want to debug. For example:

(gdb) b SimpleFileAdapter_create_connection

Function "SimpleFileAdapter_create_connection" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y
Breakpoint 1 (SimpleFileAdapter_create_connection) pending.

5. Execute Routing Service and debug your adapter.

(gdb) r -cfgFile simple_file_adapter.xml -cfgName file_to_file

8.3.5Java Adapter API

This section does not intend to give complete information on the entire Java API, but rather to describe the aspects of the Java API that are specific to the Java language.

For detailed information about the Java API, please see the online (HTML) Routing Service documentation.

The Java Adapter API defines the interfaces in Table 8.10.

8.3.5.1Adapter Entry Point

Every Java adapter must create an Adapter class that implements the com.rti.routingservice.adapter.Adapter interface.

Adapter classes are registered with Routing Service using the tag <class_name> within <java_adapter_plugin> (see Section 8.1).

When Routing Service creates a new adapter object it will look for the following constructor:

MyAdapter(java.utils.Properties properties)

If the constructor does not exist, Routing Service will use the default constructor without arguments.

MyAdapter()

8-35

Table 8.10 Java Adapter API Interfaces

Interface

 

Description

 

 

 

 

 

Required

com.rti.routingservice.adapter.Adapter

The Adapter interface defines methods to:

get the adapter version

 

 

create/destroy connections

 

 

 

Required

 

The Connection interface defines methods to:

com.rti.routingservice.adapter.Connection

create/destroy Sessions

create/destroy StreamReaders

 

 

create/destroy StreamWriters

 

update the Connection configuration

 

 

com.rti.routingservice.adapter.

The DiscoveryConnection interface defines methods to:

get the discovery StreamReaders (see Section 8.2.2) copy/

DiscoveryConnection

 

delete TypeRepresentations

 

 

 

 

 

Required

com.rti.routingservice.adapter.Session

The Session interface defines methods to update the Session

 

configuration

 

 

 

Required for input adapters

 

The StreamReader interface defines methods to:

com.rti.routingservice.adapter.StreamReader

read samples

 

return the loan on the read samples

 

update the StreamReader configuration

 

 

 

Required for output adapters

com.rti.routingservice.adapter.StreamWriter

The StreamWriter interface defines methods to:

 

write samples

 

update the StreamWriter configuration

 

 

 

8.3.5.2Error Notification

Routing Service must be notified about errors in the adapter’s logic. To do so, use the following exception: com.rti.routingservice.adapter.infrastructure.AdapterException

8.3.5.3Adapter Verbosity

The property rti.routingservice.verbosity provided to the Adapter constructor can be used to get the verbosity level used to run Routing Service.

Table 8.11 describes the mapping between the command-line option -verbosity and the values of the property "rti.routingservice.verbosity".

Table 8.11 Mapping between -verbosity and rti.routingservice.verbosity

-verbosity

rti.routingservice.verbosity

 

 

0

none

 

 

1

exception

 

 

2

warn

 

 

3 and 4

info

 

 

5 and 6

debug

 

 

8-36

8.3.6My First Java Adapter

This section shows how to create a simple Java adapter on Windows and UNIX-based architectures. It is not intended to give complete coverage of the entire adapter API, but rather to introduce the adapter technology and provide the basics of the development process of a Java adapter.

The new Adapter will be a simple file adapter where the input adapter reads lines from a text file and the output adapter saves the provided lines to an output text file.

The source code and scripts that you will create in the next sections are provided in <Routing

Service home>/adapters/tutorial/Java.

8.3.6.1Setting the Environment on the Development Machine

There are a few things to take care of before you start developing the simple file adapter.

1.Set the environment variable ROUTINGSERVICEHOME

Set the environment variable ROUTINGSERVICEHOME to the Routing Service installation directory. (Routing Service itself does not require that you set the environment variable. It is used to build, compile and run the example adapter).

2.On Windows Systems: To use a Java adapter, you must have the Visual Studio 2005

service pack 1 redistributable libraries. You can obtain this package from Microsoft or RTI (see the RTI Core Libraries and Utilities Release Notes1 for details).

3.Make sure Java 1.5 or higher is available.

Ensure that appropriate javac, jar and jdb (for debugging) executables are on your path. They can be found in the bin directory of your JDK installation.

4.Make sure you add the directory of the Java Virtual Machine dynamic library to your environment variable: LD_LIBRARY_PATH (on UNIX-based systems) or Path (on Windows systems). For example:

setenv LD_LIBRARY_PATH ${LD_LIBRARY_PATH}:/local/java/jdk1.5.0_07/jre/lib/i386/client

8.3.6.2Creating a Build Script for UNIX-based Systems

In this section, you will create a shell script to compile the Java adapter.

1.Create a directory that will contain the build script and the adapter implementation. The rest of this section assumes that you will use /opt/adapters/simplefile as the adapter directory.

2.In /opt/adapters/simplefile, create a file called build.sh with the following content.

#!/bin/sh

############################################

## RTI Routing Service File Simple Adapter ##

############################################

############################################

#Java compiler JAVAC=javac JAR=jar

#Path to RTI Routing Service Adapter API

ADAPTER_CLASSPATH="$ROUTINGSERVICEHOME/class/rtirsadapter.jar"

# Path to RTI Connext Java API DDS_CLASSPATH="$ROUTINGSERVICEHOME/class/nddsjava.jar"

ALL_SRC=`find routingservice/adapter/simplefile -name \*.java`

1. See <Connext installation directory>/ndds.<version>/doc/pdf/RTI_CoreLibrariesAndUtilities_ReleaseNotes.pdf.

8-37

mkdir -p class

# Builds all files from 'routingservice' to 'class'

echo "Building all the sources in 'rtiroutingservice' into 'class' directory..."

$JAVAC -d class -sourcepath . -classpath $ADAPTER_CLASSPATH:$DDS_CLASSPATH $ALL_SRC $JAR cf class/simplefileadapter.jar -C class routingservice

rm -rf class/routingservice

8.3.6.3Creating a Build Script for Windows Systems

In this section you will create a script to compile the Java adapter.

1.Create a directory that will contain the build script and the adapter implementation. The rest of this section assumes that you will use c:\adapters\SimpleFileAdapter as the adapter directory.

2.In c:\adapters\SimpleFileAdapter, create a file called build.cmd with the following content.

@ECHO OFF

REM #################################################

REM # RTI Routing Service Simple File Adapter #

REM #################################################

SETLOCAL enabledelayedexpansion

REM Get rid of quotes

SET ROUTINGSERVICEHOME_NQ=%ROUTINGSERVICEHOME:"=%

REM Path to Java

SET JAVAC=javac.exe

SET JAR=jar.exe

REM Path to RTI Routing Service Adapter API

SET ADAPTER_CLASSPATH="%ROUTINGSERVICEHOME_NQ%\class\rtirsadapter.jar"

REM Path to RTI Connext Java API

SET DDS_CLASSPATH="%ROUTINGSERVICEHOME_NQ%\class\nddsjava.jar"

REM Ensure 'objs' directory exists IF NOT EXIST class (

MD class

)

ECHO Building all the sources in 'routingservice' into 'class' directory...

FOR /R routingservice %%F IN (*.java) DO %JAVAC% -d class -sourcepath . -classpath "%ADAPTER_CLASSPATH%;%DDS_CLASSPATH%" "%%F"

%JAR% cf class/simplefileadapter.jar -C class routingservice RD /S /Q class\routingservice

8.3.6.4Implementing the Adapter Class

In this section you will create the adapter class for the simple file adapter.

Every Java adapter has to create an Adapter class that implements the com.rti.routingservice.adapter.Adapter interface.

Adapter classes are registered with Routing Service using the tag <class_name> within <java_adapter_plugin> (Section 8.1).

Using your favorite Java editor, create a file called SimpleFileAdapter.java under <Adapter directory>1/routingservice/adapter/simplefile.

Insert the following content:

1.c:\adapters\SimpleFileAdapter for Windows systems, or /opt/adapters/simplefile for UNIX-based systems.

8-38

/**** Simple File Adapter *****/

package routingservice.adapter.simplefile;

import com.rti.routingservice.adapter.Adapter;

import com.rti.routingservice.adapter.Connection;

import com.rti.routingservice.adapter.StreamReaderListener;

import com.rti.routingservice.adapter.infrastructure.AdapterException;

import com.rti.routingservice.adapter.infrastructure.Version;

import java.util.Properties;

/* Simple file adapter */ public class SimpleFileAdapter implements Adapter {

String verbosity;

/* Entry point to the adapter */

public SimpleFileAdapter(Properties props) { verbosity = props.getProperty(

"rti.routingservice.verbosity");

if (verbosity.equals("debug")) { System.out.println("CREATE " +

getClass().getName());

}

}

/* Create Connection */

public Connection createConnection( String routingServiceName, String routingServiceGroupName, StreamReaderListener

inputStreamDiscoveryListener, StreamReaderListener

outputStreamDiscoveryListener, Properties properties)

throws AdapterException

{

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".createConnection");

}

return new SimpleFileAdapterConnection( properties, verbosity);

}

/* Delete Connection */
public void deleteConnection(

Connection connection) throws AdapterException

{

if (verbosity.equals("debug")) { System.out.println("CALL " +

getClass().getName() + ".deleteConnection");

}

}

/* Return adapter version */ public Version getVersion() {
return new Version(1,0,0,0);

}

}

8-39

To create a SimpleFileAdapter object, Routing Service will use the constructor

SimpleFileAdapter(Properties props).

The props parameter is used to configure the adapter object. Some of the values can be set from the XML configuration file using the tag <property> within <java_adapter_plugin> and other values are set by Routing Service. One of the predefined values is "rti.routingservice.verbosity". This property provides information about the verbosity level used to run Routing Service (see Section 8.3.5.3).

Adapter objects are factories for Connection objects.

8.3.6.5Implementing the Connection Class

Connection objects provide access to data domains such as Connext domains or JMS network providers and they are configured using the XML tags <connection_1> and <connection_2> in a <domain_route> (see Section 2.4.2). In the SimpleFileAdapter example, the connection objects will provide access to a directory in your computer’s file system.

The next step consists of implementing the Connection Java class.

Create a file called SimpleFileAdapterConnection.java under <Adapter directory>1/ routingservice/adapter/simplefile.

Insert the following content:

/******* Simple File Adapter Connection *********/

package routingservice.adapter.simplefile;

import java.util.Properties;

import com.rti.routingservice.adapter.Connection;

import com.rti.routingservice.adapter.Session;

import com.rti.routingservice.adapter.StreamReader;

import com.rti.routingservice.adapter.StreamReaderListener;

import com.rti.routingservice.adapter.StreamWriter;

import com.rti.routingservice.adapter.infrastructure.AdapterException;

import com.rti.routingservice.adapter.infrastructure.StreamInfo;

/**

* Simple file connection. */

public class SimpleFileAdapterConnection implements Connection {

private String verbosity; private String directory = null;

/* Simple File Adapter Connection */ SimpleFileAdapterConnection(

Properties properties, String verbosity)

throws AdapterException

{

this.verbosity = verbosity;

directory = properties.getProperty("directory");

if (directory == null) {

1. c:\adapters\SimpleFileAdapter for Windows systems, or /opt/adapters/simplefile for UNIX-based systems

8-40

throw new AdapterException(0, "directory property is required");

}

}

/* Create Session */

public Session createSession(Properties properties) throws AdapterException

{

/* We do not need a session for the simple file adapter but we cannot return null */

return new Session() {

public void update(Properties properties) throws AdapterException {

}

};

}

/* Delete Session */

public void deleteSession(Session session) throws AdapterException

{

}

/* Create Stream Reader */

public StreamReader createStreamReader( Session session,

StreamInfo streamInfo, Properties properties, StreamReaderListener listener)

throws AdapterException

{

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".createStreamReader");

}

return new SimpleFileAdapterStreamReader( listener, streamInfo, properties, directory, verbosity);

}

/* Delete Stream Reader */
public void deleteStreamReader(

StreamReader streamReader) throws AdapterException

{

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".deleteStreamReader");

}

((SimpleFileAdapterStreamReader)streamReader).close();

}

/* Create Stream Writer */

public StreamWriter createStreamWriter( Session session,

StreamInfo streamInfo, Properties properties)
throws AdapterException

{

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".createStreamWriter");

8-41

}

return new SimpleFileAdapterStreamWriter( streamInfo, properties, directory, verbosity);

}

/* Delete Stream Writer */ public void

deleteStreamWriter(StreamWriter streamWriter) throws AdapterException

{

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".deleteStreamWriter");

}

((SimpleFileAdapterStreamWriter)streamWriter).close();

}

/* Get Attributes */

public Properties getAttributes() throws AdapterException

{

throw new AdapterException(0, "operation not supported");

}

/* Update */

public void update(Properties properties) throws AdapterException

{

}

}

Connection objects are configurable using properties (name/value pairs). The properties are set using the <property> tag within <connection_x>. For the SimpleFileAdapter example, there is one property called directory that is used to specify the directory containing the files to read/ write.

For example:

<connection_1 plugin_name="adapters::simple_file"> <registered_type name="TextLine"

type_name="TextLine"/> <property>

<value>

<element>

<name>directory</name>

<value>

/tmp/SimpleFileAdapter/input

</value>

</element>

</value>

</property> </connection_1>

Connection objects are factories for Session, StreamReader and StreamWriter objects. In the next sections you will implement StreamReader and StreamWriters. Session objects are not used in this example.

8.3.6.6Implementing the StreamReader Class

A StreamReader provides a way to read data samples of a specific type from a data domain.

8-42

In the configuration file, StreamReaders are associated with the tag <input> within <route> or <auto_route> (see Section 2.4.6).

The StreamReaders created by the SimpleFileAdapter connections read text files from the connection directory.

The data samples provided to Routing Service (using the read operation) are DynamicData with the following IDL type:

struct TextLine {
string<1024> value;

};

When a SimpleFileAdapter StreamReader is created, the name of the file is the input stream name with a .txt extension. The frequency at which the StreamReader notifies Routing Service of new lines is configurable using the read_period property. For example:

<route name="route"> <input connection="1">

<stream_name> HelloWorld </stream_name>

<registered_type_name> TextLine

</registered_type_name> <property>

<value>

<element> <name>read_period</name> <value>1000</value>

</element>

</value>

</property>

</input>

...

</route>

In the above example, the input StreamReader will read the lines of a file called HelloWorld.txt and provide one line per second to Routing Service.

The next step consist on the implementation of the StreamReader class. There are three main methods:

read()

This method will be called by Routing Service after being notified that the are new lines available. Although the signature of the method allows returning more than one sample (line), for the sake of simplicity, the implementation only returns one line every time the method is called.

Routing Service will not call the read operation until it is notified of the presence of new data (see Section 8.2.3). To provide data notification, the StreamReader implementation creates a thread (NotificationThread) that wakes up after read_period and notifies Rout- ing Service of new data if the end of the file has not been reached yet.

return_loan()

The loan on the samples provided by read() is returned to the StreamReader using this method. The SimpleFileAdapter implementation of return_loan is empty because of these reasons:

The read operation does not create new samples and it always return a single sample stored in the StreamReader.

Two calls to read() cannot occur in parallel.

8-43

update()

The update methods will be called when the read_period is changed using remote administration.

Create a file called SimpleFileAdapterStreamReader.java under <Adapter directory>1/ routingservice/adapter/simplefile.

Insert the following content:

/** Simple File Adapter Stream Reader **/
package routingservice.adapter.simplefile;

import java.io.File;

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException; import java.util.List;

import java.util.Properties;

import com.rti.dds.dynamicdata.DynamicData;

import com.rti.dds.typecode.TypeCode;

import com.rti.routingservice.adapter.StreamReader;

import com.rti.routingservice.adapter.StreamReaderListener;

import com.rti.routingservice.adapter.infrastructure.AdapterException;

import com.rti.routingservice.adapter.infrastructure.StreamInfo;

public class SimpleFileAdapterStreamReader implements StreamReader {

private String verbosity; private int readPeriod;
private String fileName = null;

private BufferedReader fileReader = null;

private NotificationThread notificationThread = null;
private DynamicData dynamicData = null;

/* Parse Properties */ private void

parseProperties(Properties properties)

{

String readPeriodStr;
readPeriodStr =

properties.getProperty("read_period");

if (readPeriodStr == null) {
readPeriod = 1000;

} else {

readPeriod = new Integer(readPeriodStr).intValue();

}

}

/* Simple File Adapter Stream Reader */ SimpleFileAdapterStreamReader(

StreamReaderListener listener,

1. c:\adapters\SimpleFileAdapter for Windows systems, or /opt/adapters/simplefile for UNIX-based systems.

8-44

StreamInfo streamInfo, Properties properties, String directory, String verbosity)

throws AdapterException

{

this.verbosity = verbosity; parseProperties(properties);

fileName = streamInfo.getStreamName() + ".txt";

try { fileReader =

new BufferedReader(new FileReader(
new File(directory,fileName)));

} catch (IOException e) {

throw new AdapterException(0, "error opening " + fileName);

}

dynamicData = new DynamicData( (TypeCode)

streamInfo.getTypeInfo().getTypeRepresentation(), DynamicData.PROPERTY_DEFAULT);

notificationThread = new NotificationThread( this, listener, fileReader, readPeriod);
notificationThread.start();

}

/* Close */ void close()

throws AdapterException {

try { notificationThread.terminate();
notificationThread.join();

if (fileReader != null) {
fileReader.close();

}

}catch (InterruptedException e) { throw new AdapterException(0,

"error finishing notification thread");

}catch (IOException e) {

throw new AdapterException(0, "error closing " + fileName);

}

}

/* Read */

public void read(List<Object> sampleList, List<Object> infoList)

throws AdapterException

{

String line;

if (verbosity.equals("debug")) {
System.out.println("CALL " +

getClass().getName() + ".read");

}

try {

8-45

sampleList.clear();

infoList.clear();
dynamicData.clear_all_members();

if (fileReader.ready()) {

line = fileReader.readLine();
dynamicData.set_string("value",

DynamicData.MEMBER_ID_UNSPECIFIED, line);

sampleList.add(dynamicData);

}

} catch (IOException e) {

throw new AdapterException(0, "error reading from file " + fileName, e);

} catch (Exception e) {

throw new AdapterException(0, "error reading", e);

}

}

/* Return Loan */
public void returnLoan(

List<Object> sampleList, List<Object> infoList) throws AdapterException {

if (verbosity.equals("debug")) { System.out.println(

"CALL " + getClass().getName() + ".returnLoan");

}

}

/* Update */

public void update(Properties properties) throws AdapterException {

parseProperties(properties);

notificationThread.setReadPeriod(readPeriod);

}

/* Notification thread

* This thread will notify of data availability in the file. */

class NotificationThread extends Thread {

private BufferedReader fileReader = null;
private int notificationPeriod;

private boolean _terminate;

private StreamReaderListener listener = null;
private StreamReader streamReader = null;

/* Notification Thread */
NotificationThread (

StreamReader streamReader, StreamReaderListener listener, BufferedReader fileReader, int notificationPeriod) {

this.listener = listener; this.fileReader = fileReader;

this.notificationPeriod = notificationPeriod;
this.streamReader = streamReader;

_terminate = false;

}

8-46

/* Run */

public void run() {
while (!_terminate) {

try { Thread.sleep(notificationPeriod);
if (fileReader.ready()) {

listener.onDataAvailable(streamReader);

}

} catch (Exception e) {}

}

}

/* Terminate */

public void terminate() { _terminate = true;

}

/* Set Read Period */

public void setReadPeriod(int readPeriod) {
notificationPeriod = readPeriod;

}

}

}

8.3.6.7Implementing the StreamWriter Class

A StreamWriter provides a way to write samples of a specific type into a data domain.

In the configuration file, StreamWriters are associated with the tag <output> within <route> or <auto_route> (see Section 2.4.6).

The SimpleFileAdapter StreamWriters create new files in the connection directory and store the lines read from the routes’ inputs.

The data samples provided to the StreamWriters’ write operation are DynamicData with the following IDL type:

struct TextLine { string<1024> value;

};

When a SimpleFileAdapter StreamWriter is created, the name of the file is the output stream name with a .txt extension. For debugging purposes, the StreamWriter can be configured to print the written samples on the console:

<route name="route">

...

<output> <stream_name>

HelloWorld </stream_name> <registered_type_name>

TextLine </registered_type_name> <property>

<value>

<element>

<name> print_to_stdout

</name>

<value>1</value>

</element>

</value>

</property>

</output>

</route>

8-47

In the above example, the output StreamWriter will store the lines provided by Routing Service on a file called HelloWorld.txt. It will also print the lines on the screen.

Insert the following code in the “Simple File Adapter: StreamWriter“ section of

SimpleFileAdapter.c.

Create a file called SimpleFileAdapterStreamWriter.java under <Adapter directory>1/ routingservice/adapter/simplefile.

Insert the following content:

/** Simple File Adapter Stream Writer **/
package routingservice.adapter.simplefile;

import java.io.File;
import java.io.FileWriter;

import java.io.BufferedWriter;
import java.io.IOException;
import java.util.List;

import java.util.ListIterator;
import java.util.Properties;

import com.rti.dds.dynamicdata.DynamicData;

import com.rti.routingservice.adapter.StreamWriter;

import com.rti.routingservice.adapter.infrastructure.AdapterException;

import com.rti.routingservice.adapter.infrastructure.StreamInfo;

public class SimpleFileAdapterStreamWriter implements StreamWriter {

private String verbosity = null; private String fileName = null;
private boolean printToStdout;

private BufferedWriter fileWriter = null;

/* Parses Properties */
private void

parseProperties(Properties properties) {

int printToStdoutInt; String printToStdoutStr;

printToStdoutStr = properties.getProperty("print_to_stdout");

if (printToStdoutStr == null) {
printToStdout = false;

}else {
printToStdoutInt =

new Integer(printToStdoutStr).intValue();

if (printToStdoutInt != 0) {
printToStdout = true;

} else {

printToStdout = false;

}

}

}

/* Simple File Adapter Stream Writer */

1. c:\adapters\SimpleFileAdapter for Windows systems, or /opt/adapters/simplefile for UNIX-based systems.

8-48

SimpleFileAdapterStreamWriter( StreamInfo streamInfo, Properties properties,

String directory, String verbosity)

throws AdapterException {

this.verbosity = verbosity;
parseProperties(properties);

fileName = streamInfo.getStreamName() + ".txt";

try {

fileWriter = new BufferedWriter ( new FileWriter (

new File (directory, fileName))); }
catch (IOException e) {

throw new AdapterException ( 0, "error opening " + fileName);

}

}

/* Adapter Exception */
void close()

throws AdapterException {

try {

if (fileWriter != null) {
fileWriter.close();

}

} catch (IOException e) {

throw new AdapterException ( 0, "error closing " + fileName);

}

}

/* Write */
public int write(

List<Object> sampleList, List<Object> infoList) throws AdapterException {

String line;

ListIterator iterator = sampleList.listIterator();
DynamicData dynamicData = null;

if (verbosity.equals("debug")) {
System.out.println ("CALL " +

getClass().getName() + ".write");

}

try {

while (iterator.hasNext()) {

dynamicData = (DynamicData) iterator.next();

line = dynamicData.get_string("value", DynamicData.MEMBER_ID_UNSPECIFIED);

fileWriter.write(line);

fileWriter.newLine();

if (printToStdout) {
System.out.println(line);

}

}

} catch (IOException e) {

throw new AdapterException(0, "error writing to file " +

8-49

fileName, e);

} catch (Exception e) {

throw new AdapterException(0, "error writing", e);

}

return 0;

}

/* Update */

public void update (Properties properties) throws AdapterException {

parseProperties(properties);

}

}

8.3.6.8Running the SimpleFileAdapter

This section describes the steps required to use and run the SimpleFileAdapter with Routing Service. You will create a configuration file with a single route that reads a HelloWorld text file from an input directory and saves it into an output directory.

1.Compile and build the SimpleFileAdapter.

UNIX-based systems:

>cd /opt/adapters/simplefile

>./build.sh

Windows systems:

>cd c:\adapters\SimpleFileAdapter

>build.cmd

2.In the adapter project directory (c:\adapters\SimpleFileAdapter1 on Windows systems; /opt/adapters/simplefile1 on UNIX-based systems), create two directories called input and output.

3.In the input directory create a file called HelloWorld.txt with the following content.

Hello World 1!

Hello World 2!

Hello World 3!

Hello World 4!

Hello World 5!

Hello World 6!

Hello World 7!

Hello World 8!

Hello World 9!

Hello World 10!

4.In the adapter project directory, create a Routing Service XML configuration file called simple_file_adapter.xml with the following content. Replace the value of the directory property under both connections with the location of the input and output directories.

<?xml version="1.0"?> <dds>

<adapter_library name="adapters"> <java_adapter_plugin name="simple_file">

<class_name> routingservice.adapter.simplefile.SimpleFileAdapter

</class_name> </java_adapter_plugin>

</adapter_library>

<types>

<struct name="TextLine"> <member name=

1. Your directory may be different if you did not use the default locations.

8-50

"value" type="string" stringMaxLength="2048"/> </struct>

</types>

<routing_service name="file_to_file"> <jvm>

<class_path> <element>

./class/simplefileadapter.jar

</element> </class_path>

</jvm>

<domain_route name="domain_route">

<connection_1 plugin_name="adapters::simple_file"> <registered_type name= "TextLine" type_name="TextLine"/>

<property>

<value>

<element>

<name>directory</name>

<value>

/opt/adapters/simplefile/input

</value>

</element>

</value>

</property> </connection_1>

<connection_2 plugin_name="adapters::simple_file"> <registered_type name="TextLine" type_name="TextLine"/>

<property>

<value>

<element>

<name>directory</name>

<value>

/opt/adapters/simplefile/output

</value>

</element>

</value>

</property> </connection_2>

<session name="session"> <route name="route">

<input connection="1"> <stream_name> HelloWorld </stream_name>

<registered_type_name> TextLine

</registered_type_name> </input>

<output> <stream_name>

HelloWorld </stream_name> <registered_type_name>

TextLine </registered_type_name> <property>

<value>

<element>

<name> print_to_stdout

</name>

<value>1</value>

8-51

</element>

</value>

</property>

</output>

</route>

</session> </domain_route>

</routing_service> </dds>

5.Start Routing Service by entering the following in a command shell.

On UNIX-based systems:

>cd <SimpleFileAdapter project directory>

>$ROUTINGSERVICEHOME/scripts/rtiroutingservice

-cfgFile simple_file_adapter.xml -cfgName file_to_file

On Windows systems:

>cd <SimpleFileAdapter project directory>

>%ROUTINGSERVICEHOME%\scripts\rtiroutingservice -cfgFile simple_file_adapter.xml -cfgName file_to_file

6.On the screen you will see:

RTI Routing Service <version> started (with name file_to_file)

Hello World 1!

Hello World 2!

Hello World 3!

Hello World 4!

Hello World 5!

Hello World 6!

Hello World 7!

Hello World 8!

Hello World 9!

Hello World 10!

7.Verify that a file called HelloWorld.txt has been generated into the output directory. The content of this file should be identical to the content of the same file in the input directory.

8.3.7Debugging Java Adapters

When you develop a custom adapter, you will need to debug it and test it. This section describes the tools and APIs that you have available to debug and detect problems in Routing Service adapters written in Java.

The first debugging capability is provided by the Routing Service Adapter SDK. The adapter SDK provides a way to access the verbosity level of Routing Service as a property called rti.routingservice.verbosity, which can be obtained from the properties passed to the adapter constructor. It is highly recommended that, as part of the adapter implementation, you instrument the code by adding status messages that will be printed with the INFO and DEBUG verbosity levels. This level of instrumentation will help you to capture run-time information for troubleshooting.

The second debugging capability is provided by third party tools. The rest of this section shows how to debug a Java adapter using jdb (the command-line debugger of Java) and NetBeans™ IDE 6.9.

8.3.7.1Enabling Debugging in the Routing Service JVM

Before you start debugging with jdb or NetBeans, you have to enable debugging in the JVM- created Routing Service.

8-52

1.If you have not done so already, stop the existing Routing Service execution by pressing CTRL-C.

2.Edit java_simple_adapter.xml and replace the content of the JVM tag with:

<jvm> <class_path>

<element>

./class/simplefileadapter.jar

</element> </class_path> <options>

<element>-Xdebug</element> <element>

-Xrunjdwp:transport=dt_socket,address=8192,server=y,suspend=y
</element>

</options>

</jvm>

The JVM option -Xdebug is used to enable debugging.

The JVM option -Xrunjdwp loads the JDPA reference implementation of JDWP (Java Debug Wire Protocol) and starts listening on port 8192 to communicate with a separate debugger application such as jdb and NetBeans.

For additional details on Java debugging see:

http://java.sun.com/javase/technologies/core/toolsapis/jpda

3.Save the changes.

4.Run Routing Service.

On UNIX-based systems:

>cd <SimpleFileAdapter project directory>

>$ROUTINGSERVICEHOME/scripts/rtiroutingservice

-cfgFile simple_file_adapter.xml -cfgName file_to_file

On Windows systems:

>cd <SimpleFileAdapter project directory>

>%ROUTINGSERVICEHOME%\scripts\rtiroutingservice -cfgFile simple_file_adapter.xml -cfgName file_to_file

You should see output like this:

Listening for transport dt_socket at address: 1024

At this point, the execution of Routing Service is suspended and waiting for a debugger to attach.

8.3.7.2Debugging with JDB

jdb is the command-line debugger of Java. This section is not intended to give complete coverage on all the jdb functionality and commands, but rather to provide basic information on how to attach to the Routing Service JVM and start debugging.

For more information about JDB see the following web page: http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/jdb.html

1.In a separate shell window, start jdb:

UNIX-based systems:

>cd /opt/adapters/simplefile

>jdb -sourcepath . -attach 1024

Windows systems:

>cd c:\adapters\SimpleFileAdapter

>jdb -sourcepath . -attach 1024

8-53

2. Set breakpoints in the methods or classes that you would like to debug.

For example, to set a breakpoint in the SimpleFileAdapter constructor enter the following:

main[1] stop in routingservice.adapter.simplefile.SimpleFileAdapter.<init>

3. Resume the execution of Routing Service by entering:

main[1] cont

You will see output similar to:

> Set deferred breakpoint routingservice.adapter.simplefile.SimpleFileAdapter.<init>
Breakpoint hit: "thread=main",
routingservice.adapter.simplefile.SimpleFileAdapter.<init>(), line=23 bci=0

23 public SimpleFileAdapter(Properties props) {

Use the command help to get a list of the command that will allow you to continue the debugging process.

8.3.7.3Debugging with NetBeans

NetBeans is an IDE for developing and debugging Java applications. This section is not intended to give complete coverage of all the NetBeans debugger functionality, but rather to provide basic information on how to attach the NetBeans debugger to the Routing Service JVM and start debugging.

1.Verify that NetBeans IDE 6.9 is installed on your system. The installation of NetBeans is beyond the scope of this document; please refer to NetBeans documentation.

2.Start NetBeans.

3.Make the adapter source code available to the debugger.

a.Select Window, Debugging, Sources.

b.Right-click the Remote Attach window and select Add Source Root.

c.Enter the adapter directory.

4.Set breakpoints in the methods or classes that you would like to debug.

For example, to set a breakpoint in the SimpleFileAdapter constructor, follow the following steps:

a. Select Debug, New Breakpoint.

8-54

b.In the New Breakpoint window, select Method as the breakpoint type and provide routingservice.adapter.simplefile.SimpleFileAdapter as the class name and <init> as the method name.

c.Press OK.

5.Attach the debugger to Routing Service JVM.

a.Select Debug, Attach Debugger.

b.For the Host, enter the name of the host where Routing Service is running.

c.For the Port, enter 8192.

d.Press OK to start debugging the adapter.

8.3.8Testing an Adapter

A simple Java test adapter is provided with Routing Service Adapter SDK. You will find the class, com.rti.routingservice.adapter.test.TestAdapter, in rtirsadapter.jar.

This is a convenient way to test your own adapters. The TestAdapter is used as an output adapter that counts the number of samples that meet certain conditions defined in the configuration file.

8-55

Your adapter will act as the input and its samples will be passed to the TestAdapter. If the number of samples received by the TestAdapter is not between a defined range when you stop Routing Service, you will see a failure message. (Success or failure is determined when you stop Routing Service and it destroys the adapter.)

To use the TestAdapter to test your input adapter:

1.Write a configuration file in which your adapter is the input for one or more routes and the TestAdapter is the output.

Configure the TestAdapter with the expected number of samples within a range specified using the properties MinExpectedSamples and MaxExpectedSamples in the <output> tag.

2.Run Routing Service using that configuration file.

3.Wait the amount of time your adapter may require.

4.Stop Routing Service. The TestAdapter will print a failure or success message.

You can avoid steps 3-4 by starting Routing Service with the -stopAfter <seconds> command- line option.

If you run Routing Service with -verbosity 3 (or higher), the TestAdapter will also print the DynamicData samples as they arrive.

For an example of how to use and configure the TestAdapter, see <Routing Service home>/ example/testing/test_adapter.xml. This example tests the simple C file adapter introduced in previous sections.

You can also write your own adapter to extend the TestAdapter class. The source code is in rtirsadapter.jar.

8-56