By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress.”
The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html.
This Internet-Draft will expire on April 17, 2009.
This document defines the format of Uniform Resource Identifiers (URI) as defined in [RFC3986] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.), for designating connections and destination addresses used in the Java(tm) Messaging Service (JMS) [REF‑JMS] (Hapner, M., Burridge, R., Sharma, R., Fialli, J., and K. Stout, “Java Message Service (JMS),” April 2002.). It was originally designed for particular uses, but should have general applicability wherever a JMS URI is needed to describe the connection to a JMS provider, and access to a JMS destination. The syntax of this 'jms' URI is not compatible with any known current vendor implementation, but the expressivity of the format should permit all vendors to use it.
1.1. Requirements notation
2. URI Scheme Name
3. Syntax of a jms URI
4. URI scheme semantics
4.1. Shared Parameters
4.2. JNDI Variant
4.3. Vendor Destination Names - Variants "queue" And "topic"
4.4. Custom parameters
5. Encoding considerations
6. Applications/protocols that use the JMS URI scheme name
7. Interoperability considerations
8. Security Considerations
8.1. Reliability and Consistency
8.2. Malicious Construction
8.3. Back-end Transcoding
8.4. Semantic Attacks
8.5. Other Security Concerns
9. IANA Considerations
12.1. Normative References
12.2. Informative References
§ Authors' Addresses
§ Intellectual Property and Copyright Statements
The "jms" URI scheme is used to designate a javax.jms.Destination object and an associated javax.jms.ConnectionFactory object, and optionally provide additional information concerning the way that the Destination object is to be used. Probably the most common, and certainly the most compatible way in Java to retrieve such destinations, is via Java Naming and Directory Information (JNDI) [REF‑JNDI] (Sun Microsystems, Inc., “Java Naming and Directory Interface Application Programming Interface,” July 1999.) methods. So as to extend compatibility to existing vendor mechanisms beyond JNDI lookup, the "jms" URI syntax allows variants on the core syntax. The variant exists as an explicit part of the syntax so that tools that are otherwise unfamiliar with the variant can recognize the presence of a URI with an alternate interpretation.
In its simplest and most interoperable form, this URI scheme starts with "jms:jndi:" plus a JNDI name for a Destination. Since interaction with some resources may require JNDI contextual information or JMS headers and properties to be specified as well, the "jndi" variant of the "jms" URI scheme includes support for supplying this additional JNDI information as query parameters.
While the "jndi" variant provides compatibility, vendors may define additional variants. This specification defines three variants, "jndi", "queue", and "topic".
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119] (Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” March 1997.).
All syntax descriptions use the ABNF specified by [RFC5234] (Crocker, D. and P. Overell, “Augmented BNF for Syntax Specifications: ABNF,” January 2008.), Augmented BNF for Syntax Specifications: ABNF.
The name of the URI scheme is 'jms'.
The following ABNF describes the jms scheme URI syntax:
jms-uri = "jms:" jms-variant ":" jms-dest [ "?" param *( "&" param ) ] jms-variant = segment-nz-nc jms-dest = path-rootless ; specific meaning per variant param = param-name "=" param-value param-name = 1*(unreserved / pct-encoded) param-value = *(unreserved / pct-encoded) segment-nz-nc = <as defined in RFC 3986> path-rootless = <as defined in RFC 3986> unreserved = <as defined in RFC 3986> pct-encoded = <as defined in RFC 3986>
JMS URI schemes are used to locate JMS Destination resources and do not specify actions to be taken on those resources. Operations available on JMS destinations are fully and normatively defined by the JMS specification and as such, are out of scope for this URI specification.
The required particles in the JMS URI are the scheme name ("jms"), the variant identifier, and the "jms-dest" portions. The three recognized variants (jms-variant above) are "jndi", "queue", and "topic". The "jms-dest" portion identifies the JMS destination object in a way that is determined by the particular variant.
Each variant may have query parameters specific to that variation. All such parameters that cannot be shared across schemes should use the name of the variant as the prefix to the parameters. Parameters that apply across multiple variants, perhaps because they are generally applicable, such as JMS settings, should not have any particular prefix, and should not begin with any known prefix. This latter convention enables tools that are otherwise unfamiliar with a particular variant to recognize that a particular URI includes parameters specific to that variant.
Examples of the URI scheme include:
jms:jndi:SomeJndiNameForDestination?jndiInitialContextFactory= com.example.jndi.JndiFactory&priority=3 jms:queue:ExampleQueueName?timeToLive=1000
In addition to the required particles, the jms URI scheme supports the following "shared" parameters, which may be included as parameters in any order (following the '?' parameter-start indicator, and separated by '&'). This pattern is consistent with other non-hierarchical URI specifications.
Indicates whether the request message is persistent or not. This property corresponds to the JMS message header "JMSDeliveryMode" defined in section 3.4.2. of the JMS 1.1 specification. This may be "PERSISTENT" or "NON_PERSISTENT". If this parameter is not specified then the JMS default SHOULD be used.
The lifetime, in milliseconds, of the request message. This property corresponds to the JMS Time-To-Live value defined in section 4.8 of the JMS 1.1 specification. If this parameter is not specified then the JMS default SHOULD be used.
The JMS priority associated with the request message. As per section 3.4.10 of the JMS 1.1 specification this must be a number between 0 and 9, inclusive, and corresponds to the JMS message header field "JMSPriority". If this parameter is not specified then the JMS default SHOULD be used.
This property corresponds to the JMS message header "JMSReplyTo" defined in section 3.4.6 of the JMS 1.1 specification. Specifies the JMS destination object to which a response message should be sent in a way that is determined by the particular variant.
The "jndi" variant implies the use of JNDI for discovering the Destination object. When this is specified as the variant, the jms-dest portion of the syntax is the name for JNDI lookup purposes. Additional JNDI specific parameters may be specified. The JNDI specific parameters SHOULD only be processed when the URI variant is "jndi".
Specifies the JNDI name of the Java class providing the connection factory.
Specifies the fully qualified Java class name of the "InitialContextFactory" implementation class to use.
Specifies the JNDI provider URL, in a form consistent with javax.naming.spi.NamingManager.getURLContext(String scheme, Hashtable environment) as defined in the JNDI specification.
To perform a look-up based on a JNDI variant URI an application must create a JNDI InitialContext object. The InitialContext object can then be used to look up the JMS ConnectionFactory object (using the "jndiConnectionFactoryName" URI parameter); the target JMS Destination object (using the "jms-dest" portion of the JMS URI); and the "replyToName" JMS Destination object (if the "replyToName" parameter is specified on the URI).
The application creates the InitialContext object by first setting up two properties: "Context.INITIAL_CONTEXT_FACTORY", with the value of the jndiInitialContextFactory JMS URI parameter; and "Context.PROVIDER_URL", with the value of the jndiURL URI parameter, and then passing the two properties to the InitialContext constructor.
To locate a connection factory or destination object, the application passes the name of the object into the InitialContext.lookup() method.
For example, the JMS URI...
jms:jndi:REQ_QUEUE?jndiURL=file:/C:/JMSAdmin &jndiInitialContextFactory=com.sun.jndi.fscontext.RefFSContextFactory &jndiConnectionFactoryName=CONNFACT &replyToName=RESP_QUEUE
...would be used by the following (non-normative) code sample to locate and retrieve a JMS ConnectionFactory called "CONNFACT", and JMS Destinations called "REQ_QUEUE" and "RESP_QUEUE", from a file system JNDI context called "c:/JMSAdmin".
/* * Preconditions on URI: * - portion "jms-dest" has been parsed into variable "jms_dest" * - parameters "jndiConnectionFactoryName", * "jndiInitialContextFactory", "replyToName" and "jndiURL" have * been parsed into variables of the same name */ Hashtable environment = new Hashtable(); environment.put(Context.INITIAL_CONTEXT_FACTORY, jndiInitialContextFactory); environment.put(Context.PROVIDER_URL, jndiURL); /* * Create File System Initial Context */ Context ctx = new InitialContext(environment); /* * Now get the JMS ConnectionFactory and Destination. These will be * used later on in the application to create the JMS Connection and * send / receive messages */ ConnectionFactory jmsConnFact = (ConnectionFactory) ctx.lookup(jndiConnectionFactoryName); Destination requestDest = (Destination) ctx.lookup(jms_dest); Destination replyDest = (Destination) ctx.lookup(replyToName);
The ConnectionFactory is used to create a Connection, which itself is used to create a Session. The session can then be used to create the MessageProducer - which sends messages to the target destination, and the MessageConsumer which receives messages from the replyToName destination (as shown in the following code extract)
/* * Create a producer to send a message to the request destination * that was specified in the URI, then create the message, setting * the replyToName destination in the message to the one specified * in the URI, and send it. */ MessageProducer producer = sess.createProducer(requestDest); BytesMessage reqMsg = sess.createBytesMessage(); reqMsg.setJMSReplyTo(replyDest); producer.send(reqMsg); /* * Create a consumer to get a message from the replyToName * destination using a selector to get the specific response to this * request. The responder must set the correlation ID of the response * to the message ID of the request message */ MessageConsumer consumer = sess.createConsumer(replyDest, "JMSCorrelationID = '" + reqMsg.getJMSMessageID() + "'"); Message respMsg = (Message) consumer.receive(300000);
The JMS Session object provides a means to directly access queues and topics. Specifically, it has the methods Session.createQueue(String name), and Session.createTopic(String name). These methods can be used to "create" the Java representation of an existing JMS Topic or Queue.
Since the Session interface requires external knowledge about whether a given name relates to a queue or topic, rather than introducing one new variant, this section defines two variants. A JMS URI can indicate which of these methods to use by specifying the appropriate variant - either "queue" or "topic". For example:
to identify a JMS queue Destination, and
to identify a JMS topic Destination.
JMS only specifies one way to obtain the names used by these APIs. With a JMS Queue or Topic available, an implementation can call Queue.getQueueName(), or Topic.getTopicName(), respectively, both of which return a String object. To create a correct corresponding URI, the resulting string must use standard URI escape mechanisms so that the resulting characters conform to "jms-dest".
When used with the "queue" and "topic" variants, the replyToName parameter, specified in section 4.1.4, always refers to a name of a JMS queue to look up via the Session.createQueue() method. For either variant, if a JMS topic is instead required as a response destination, a JMS URI can emply the "topicReplyToName" parameter. This parameter defines a name to look up with the Session.createTopic() method.
A JMS URI MUST NOT specify both a "topicReplyToName" and a "replyToName" paramter.
Using the Session.createQueue(), and Session.createTopic() methods assumes that a client program has already obtained a Session object. Where does that Session object come from - how does a client get it?
One way to get a Session object is to simply revert to using JNDI. That is, if a Session is not available to client from some other context, the "queue" and "topic" variants MAY reuse the URL parameters specified in section 4.2.1, JNDI Parameters. Via JNDI, those parameters will identify a ConnectionFactory, which can then be used to obtain a Session object.
Combining the "queue" and "topic" variants with JNDI lookup for an implementation of ConnectionFactory raises an important consideration for JMS URI clients. Once clients are using JNDI for one part of discovering a Destination, they almost certainly could use a vendor-neutral JNDI lookup for a Destination object, rather than a vendor-specific means. As a result, clients should carefully consider alternatives to this approach.
The JMS specification clearly identifies the two methods on the Session interface as returning vendor specific names for destinations. Consequently, users of the JMS URI scheme should carefully consider when these two variants should be applied. If users plan switching between JMS vendors, they should also plan on regenerating resources that contain URIs in this vendor specific form.
A JMS vendor may provide alternate ways to obtain the names that can be passed to Session.createQueue(), and Session.createTopic(). When using those alternate means, users of this URI specification should verify that the obtained names work as expected in all circumstances.
The set of parameters is extensible. Any other vendor- or application-defined parameter may be supplied, in the URI, by passing it as "param-name=param-value" just like the set of well-known parameters.
Warning: Vendors and applications MUST NOT include sensitive information (such as authorization tokens) in a URI. Other means of authorization, authentication, and identification should be used. Also see the security discussion below about properties that may be duplicated as JMS message properties.
The jms URI scheme distinguishes between "unreserved" characters and "pct-encoded" characters, as defined in [RFC3986] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.). Apart from these encoding considerations, the characters '?' and '&' MUST be encoded when they appear within the "jms-dest" particle (for example, a JNDI name) or in query parameters. The character ':' SHOULD be escaped, when appearing in the "jms-dest" portion of the syntax.
Conversions to and from IRIs should follow the rules of RFC 3987, sections 3.1 and 3.2. As per sections 1.2c and 6.4 of [RFC3987] (Duerst, M. and M. Suignard, “Internationalized Resource Identifiers (IRIs),” January 2005.), all parts of the jms URI MUST use the UTF-8 encoding when converting to and from IRI format.
A variety of vendors provide implementations of the JMS Service Provider Interface. These products interoperate at the API level, in the Java programming language.
Some vendors have provided additional products which interoperate with their own SPI implementations. These extensions may also be able to make use of this URI scheme.
The vendors working on this URI scheme are also working on a specification for carrying SOAP messages over their respective implementations of JMS [REF‑SOAPJMS] (Daniels, G., Easton, P., Frank, T., Johnson, E., Lewis, A., Merrick, R., Phillips, M., and D. Xiao, “SOAP over JMS,” October 2007.). In addition, the Service Component Architecture Bindings TC at OASIS will employ the jms URI scheme to identify JMS Destinations in appropriate circumstances.
This jms URI scheme focuses on identifying a JMS Destination object, and some characteristics of communication using that Destination, and specifically excludes any notion of describing how JMS itself is implemented and how it delivers messages. As a consequence of this focus, interoperability concerns are limited to how implementations obtain and use a Destination object.
This scheme definition describes two variants for obtaining a Destination. These variants achieve their aims with the use of JNDI and JMS APIs, with no new APIs or protocols defined here. As a consequence, interoperability concerns may arise as a result of implementations that do not conform to the specifications for those APIs. Further, the use of Java, and JNDI in particular, means that the configuration of the execution environment, and the use of Java ClassLoaders may affect the interpretation of any given URI. Consumers of these URIs are urged to consider the scope and consistency of the environment across which these URIs will be shared.
As described in Section 4 (URI scheme semantics), others can define additional variants, which provide the means to describe how to look up JMS Destination objects in a manner specific to some environment. For any new variant, the shared parameters defined in Section 4.1 (Shared Parameters) MUST have the same meaning in that variant as they do here. That way, tools and people can safely copy these parameters between environments. Customers should be aware that use of variants not defined here may make it more difficult to switch to an alternate JMS provider.
Section 7 of [RFC3986] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.) identifies some of the security concerns that should be identified in this specification.
This specification identifies only the variant ("jms-variant") and variant specific details ("jms-dest") as an essential part of the URI. For reliability and consistency purposes, these are the only part that can reasonably be expected to be stable. Other optional JMS configuration and message properties, indicated as URI parameters, like the "timeToLive", may reasonably be determined by the sender of a message, without affecting the recipient. Insofar as a recipient may wish to dictate certain parameters, such as the "jndiConnectionFactoryName", those parameters can be specified.
A malicious consumer of a service using a JMS URI could send, as part of a JMS message, a URI with a parameter such as "timeToLive" with a value specified in the URI that differs from the corresponding JMS message property ("JMSExpiration" header, in this example). In the case of such messages with such URIs, recipients are strongly cautioned to avoid applying processing logic based on particular URI parameters. Discrepancies in the message could be used to exploit differences in behavior between the selectors that a JMS-based application might use to affect which messages it sees, and the processing of the rest of the application. As defined in this document, the parameters of concern include:
Message senders are strongly urged to remove from the URI extra parameters like the above in environments where the data will be redundant with information specified elsewhere in the JMS message.
Any use of additional parameters, either as a part of a definition of a new variant, or for more general use, should also specify whether those parameters should be removed by a sender as specified here. If a recipient is aware of the jms URI scheme, and it receives a message containing a JMS URI, it MUST ignore or discard parameters that it does not recognize.
A third party could intercept and replace a URI containing any of the JMS/JNDI configuration parameters, such as "jndiConnectionFactoryName", "jndiInitialContextFactory", "jndiURL". As these parameters may affect how an implementation establishes an initial connection, such parameters could be used as a means to subvert communications. This could possibly result in re-routing communications to third-parties, who could then monitor sent messages. Clients should use these URI parameters only when assured of their validity in trusted environments.
This specification, in using the URI specification, and building around the JMS specification, has no particular transcoding issues. Any such issues are problems with the underlying implementation of Java and Java Messaging Service being employed.
A possible semantic attack on the "jndi" variant could be accomplished by replacing characters of the JMS URI from one language with equivalent looking characters from another language, known as an "IDN homograph attack" (IDN) (Unknown, “IDN Homograph attack,” any 2005-2006.) [REF‑Homograph]. This kind of attack could occur in a variety of ways. For example, if an environment allows for the automatic registration of JNDI destination names, a malicious actor could register and then publicize an alternate of an existing destination name. Such an environment ought to prevent the use of homograph equivalents, perhaps by restricting allowed characters, so that clients do not accidentally send their requests to unintended destinations.
The "queue" and "topic" variants are subject to the same concerns as the JNDI variant. In addition, because the destination names are vendor defined, URIs employing these two variants may employ special characters that significantly change the meaning of the URI. It is possible that the introduction of a single character - difficult for a human to notice - might dramatically change the intended meaning of a URI. In situations where this might be an issue, users of this URI should strongly consider the "jndi" variant instead.
This specification does not define or anticipate any use for IP addresses as part of the URI, so no issues around IP addresses, rare or otherwise, are raised by this specification.
This specification does not define any characteristics of a jms scheme URI that contain sensitive information.
The IANA is asked to register the Java Message Service URI scheme described in this document, according to the following scheme registration request:
The authors gratefully acknowledge the contributions of:
Phil Adams - International Business Machines Corporation - email@example.com
Glen Daniels - WSO2 - firstname.lastname@example.org
Peter Easton - Progress Software - email@example.com
Tim Frank - Software AG. - firstname.lastname@example.org
Lei Jin - BEA Systems, Inc. - email@example.com
Eric Johnson - TIBCO Software Inc. - firstname.lastname@example.org
Vinod Kumar - BEA Systems, Inc. - email@example.com
Amelia A. Lewis - TIBCO Software Inc. - firstname.lastname@example.org
Roland Merrick - International Business Machines Corporation - email@example.com
Mark Phillips - International Business Machines Corporation - firstname.lastname@example.org
Derek Rokicki - Software AG. - email@example.com
Stephen Todd - International Business Machines Corporation - firstname.lastname@example.org
Dongbo Xiao - BEA Systems, Inc. - email@example.com
Prasad Yendluri - Software AG - firstname.lastname@example.org
Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both.
This document was produced using the xml2rfc tool [RFC2629] (Rose, M., “Writing I-Ds and RFCs using XML,” June 1999.).
|[RFC2119]||Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” BCP 14, RFC 2119, March 1997 (TXT, HTML, XML).|
|[RFC3986]||Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” STD 66, RFC 3986, January 2005 (TXT, HTML, XML).|
|[RFC3987]||Duerst, M. and M. Suignard, “Internationalized Resource Identifiers (IRIs),” RFC 3987, January 2005 (TXT).|
|[RFC5234]||Crocker, D. and P. Overell, “Augmented BNF for Syntax Specifications: ABNF,” STD 68, RFC 5234, January 2008 (TXT).|
|[REF-Homograph]||Unknown, “IDN Homograph attack,” any 2005-2006.|
|[REF-JMS]||Hapner, M., Burridge, R., Sharma, R., Fialli, J., and K. Stout, “Java Message Service (JMS),” April 2002.|
|[REF-JNDI]||Sun Microsystems, Inc., “Java Naming and Directory Interface Application Programming Interface,” July 1999.|
|[REF-SOAPJMS]||Daniels, G., Easton, P., Frank, T., Johnson, E., Lewis, A., Merrick, R., Phillips, M., and D. Xiao, “SOAP over JMS,” October 2007.|
|[RFC2629]||Rose, M., “Writing I-Ds and RFCs using XML,” RFC 2629, June 1999 (TXT, HTML, XML).|
|International Business Machines Corporation|
|PO Box 31, Birmingham Road|
|Warwick, CV34 5JL|
|Progress Software Corporation|
|14 Oak Park Drive|
|Bedford, MA 01730|
|11700 Plaza America Drive|
|Reston VA 20190|
|TIBCO Software Inc.|
|3303 Hillview Avenue|
|Palo Alto CA 94304|
Copyright © The IETF Trust (2008).
This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights.
This document and the information contained herein are provided on an “AS IS” basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at email@example.com.