Message Queue administered objects are of two basic kinds:
Connection factories are used by client applications to create connections to brokers.
Destinations represent locations on a broker with which client applications can exchange (send and retrieve) messages.
Each type of administered object has certain attributes that determine the object’s properties and behavior. This section describes how to use the Object Manager command line utility (imqobjmgr) to set these attributes; you can also set them with the GUI Administration Console, as described in Working With Administered Objects.
Client applications use connection factory administered objects to create connections with which to exchange messages with a broker. A connection factory’s attributes define the properties of all connections it creates. Once a connection has been created, its properties cannot be changed; thus the only way to configure a connection’s properties is by setting the attributes of the connection factory used to create it.
Message Queue defines two classes of connection factory objects:
ConnectionFactory objects support normal messaging and nondistributed transactions.
XAConnectionFactory objects support distributed transactions.
Both classes share the same configuration attributes, which you can use to optimize resources, performance, and message throughput. These attributes are listed and described in detail in Chapter 18, Administered Object Attribute Reference and are discussed in the following sections below:
Connection handling attributes specify the broker address to which to connect and, if required, how to detect connection failure and attempt reconnection. They are summarized in Table 18–1.
The most important connection handling attribute is imqAddressList, which specifies the broker or brokers to which to establish a connection. The value of this attribute is a string containing a broker address or (in the case of a broker cluster) multiple addresses separated by commas. Broker addresses can use a variety of addressing schemes, depending on the connection service to be used (see Configuring Connection Services) and the method of establishing a connection:
mq uses the broker’s Port Mapper to assign a port dynamically for either the jms or ssljms connection service.
mqtcp bypasses the Port Mapper and connects directly to a specified port, using the jms connection service.
mqssl makes a Secure Socket Layer (SSL) connection to a specified port, using the ssljms connection service.
http makes a Hypertext Transport Protocol (HTTP) connection to a Message Queue tunnel servlet at a specified URL, using the httpjms connection service.
https makes a Secure Hypertext Transport Protocol (HTTPS) connection to a Message Queue tunnel servlet at a specified URL, using the httpsjms connection service.
These addressing schemes are summarized in Table 18–2.
The general format for each broker address is
scheme://address
where scheme is one of the addressing schemes listed above and address denotes the broker address itself. The exact syntax for specifying the address varies depending on the addressing scheme, as shown in the “Description” column of Table 18–2. Table 18–3 shows examples of the various address formats.
In a multiple-broker cluster environment, the address list can contain more than one broker address. If the first connection attempt fails, the Message Queue client runtime will attempt to connect to another address in the list, and so on until the list is exhausted. Two additional connection factory attributes control the way this is done:
imqAddressListBehavior specifies the order in which to try the specified addresses. If this attribute is set to the string PRIORITY, addresses will be tried in the order in which they appear in the address list. If the attribute value is RANDOM, the addresses will instead be tried in random order; this is useful, for instance, when many Message Queue clients are sharing the same connection factory object, to prevent them from all attempting to connect to the same broker address.
imqAddressListIterations specifies how many times to cycle through the list before giving up and reporting failure. A value of -1 denotes an unlimited number of iterations: the client runtime will keep trying until it succeeds in establishing a connection or until the end of time, whichever occurs first.
Because enhanced clusters are self-configuring (see Cluster Configuration Properties and Connecting Brokers into an Enhanced Cluster), their membership can change over time as brokers enter and leave the cluster. In this type of cluster, the value of each member broker’s imqAddressList attribute is updated dynamically so that it always reflects the cluster’s current membership.
By setting certain connection factory attributes, you can configure a client to reconnect automatically to a broker in the event of a failed connection. For standalone brokers or those belonging to a conventional broker cluster (see Conventional Clusters in Sun Java System Message Queue 4.3 Technical Overview), you enable this behavior by setting the connection factory’s imqReconnectEnabled attribute to true. The imqReconnectAttempts attribute controls the number of reconnection attempts to a given broker address; imqReconnectInterval specifies the interval, in milliseconds, to wait between attempts.
If the broker is part of a conventional cluster, the failed connection can be restored not only on the original broker, but also on a different one in the cluster. If reconnection to the original broker fails, the client runtime will try the other addresses in the connection factory’s broker address list (imqAddressList). The imqAddressListBehavior and imqAddressListIterations attributes control the order in which addresses are tried and the number of iterations through the list, as described in the preceding section. Each address is tried repeatedly at intervals of imqReconnectInterval milliseconds, up to the maximum number of attempts specified by imqReconnectAttempts.
Note, however, that in a conventional cluster, such automatic reconnection only provides connection failover and not data failover: persistent messages and other state information held by a failed or disconnected broker can be lost when the client is reconnected to a different broker instance. While attempting to reestablish a connection, Message Queue does maintain objects (such as sessions, message consumers, and message producers) provided by the client runtime. Temporary destinations are also maintained for a time when a connection fails, because clients might reconnect and access them again; after giving clients time to reconnect and use these destinations, the broker will delete them. In circumstances where the client-side state cannot be fully restored on the broker on reconnection (for instance, when using transacted sessions, which exist only for the duration of a connection), automatic reconnection will not take place and the connection’s exception handler will be called instead. It is then up to the client application to catch the exception, reconnect, and restore state.
By contrast, in an enhanced cluster (see High-Availability Clusters in Sun Java System Message Queue 4.3 Technical Overview), another broker can take over a failed broker’s persistent state and proceed to deliver its pending messages without interruption of service. In this type of cluster, automatic reconnection is always enabled. The connection factory’s imqReconnectEnabled, imqAddressList, and imqAddressListIterations attributes are ignored. The client runtime is automatically redirected to the failover broker. Because there might be a short time lag during which the failover broker takes over from the failed broker, the imqReconnectAttempts connection factory attribute should be set to a value of -1 (client runtime continues connect attempts until successful).
Automatic reconnection supports all client acknowledgment modes for message consumption. Once a connection has been reestablished, the broker will redeliver all unacknowledged messages it had previously delivered, marking them with a Redeliver flag. Client applications can use this flag to determine whether a message has already been consumed but not yet acknowledged. (In the case of nondurable subscribers, however, the broker does not hold messages once their connections have been closed. Thus any messages produced for such subscribers while the connection is down cannot be delivered after reconnection and will be lost.) Message production is blocked while automatic reconnection is in progress; message producers cannot send messages to the broker until after the connection has been reestablished.
The Message Queue client runtime can be configured to periodically test, or “ping,” a connection, allowing connection failures to be detected preemptively before an attempted message transmission fails. Such testing is particularly important for client applications that only consume messages and do not produce them, since such applications cannot otherwise detect when a connection has failed. Clients that produce messages only infrequently can also benefit from this feature.
The connection factory attribute imqPingInterval specifies the frequency, in seconds, with which to ping a connection. By default, this interval is set to 30 seconds; a value of -1 disables the ping operation.
The response to an unsuccessful ping varies from one operating-system platform to another. On some platforms, an exception is immediately thrown to the client application’s exception listener. (If the client does not have an exception listener, its next attempt to use the connection will fail.) Other platforms may continue trying to establish a connection to the broker, buffering successive pings until one succeeds or the buffer overflows.
The connection factory attributes listed in Table 18–4 support client authentication and the setting of client identifiers for durable subscribers.
All attempts to connect to a broker must be authenticated by user name and password against a user repository maintained by the message service. The connection factory attributes imqDefaultUsername and imqDefaultPassword specify a default user name and password to be used if the client does not supply them explicitly when creating a connection.
As a convenience for developers who do not wish to bother populating a user repository during application development and testing, Message Queue provides a guest user account with user name and password both equal to guest. This is also the default value for the imqDefaultUsername and imqDefaultPassword attributes, so that if they are not specified explicitly, clients can always obtain a connection under the guest account. In a production environment, access to broker connections should be restricted to users who are explicitly registered in the user repository.
The Java Message Service Specification requires that a connection provide a unique client identifier whenever the broker must maintain a persistent state on behalf of a client. Message Queue uses such client identifiers to keep track of durable subscribers to a topic destination. When a durable subscriber becomes inactive, the broker retains all incoming messages for the topic and delivers them when the subscriber becomes active again. The broker identifies the subscriber by means of its client identifier.
While it is possible for a client application to set its own client identifier programmatically using the connection object’s setClientID method, this makes it difficult to coordinate client identifiers to ensure that each is unique. It is generally better to have Message Queue automatically assign a unique identifier when creating a connection on behalf of a client. This can be done by setting the connection factory’s imqConfiguredClientID attribute to a value of the form
${u}factoryID
The characters ${u} must be the first four characters of the attribute value. (Any character other than u between the braces will cause an exception to be thrown on connection creation; in any other position, these characters have no special meaning and will be treated as plain text.) The value for factoryID is a character string uniquely associated with this connection factory object.
When creating a connection for a particular client, Message Queue will construct a client identifier by replacing the characters ${u} with ${u:userName}, where userName is the user name authenticated for the connection. This ensures that connections created by a given connection factory, although identical in all other respects, will each have their own unique client identifier. For example, if the user name is Calvin and the string specified for the connection factory’s imqConfiguredClientID attribute is ${u}Hobbes, the client identifier assigned will be ${u:Calvin}Hobbes.
This scheme will not work if two clients both attempt to obtain connections using the default user name guest, since each would have a client identifier with the same ${u} component. In this case, only the first client to request a connection will get one; the second client’s connection attempt will fail, because Message Queue cannot create two connections with the same client identifier.
Even if you specify a client identifier with imqConfiguredClientID, client applications can override this setting with the connection method setClientID. You can prevent this by setting the connection factory’s imqDisableSetClientID attribute to true. Note that for an application that uses durable subscribers, the client identifier must be set one way or the other: either administratively with imqConfiguredClientID or programmatically with setClientID.
Because “payload” messages sent and received by clients and control messages (such as broker acknowledgments) used by Message Queue itself pass over the same client-broker connection, excessive levels of payload traffic can interfere with the delivery of control messages. To help alleviate this problem, the connection factory attributes listed in Table 18–5 allow you to manage the relative flow of the two types of message. These attributes fall into four categories:
Acknowledgment timeout specifies the maximum time (imqAckTimeout) to wait for a broker acknowledgment before throwing an exception.
Connection flow metering limits the transmission of payload messages to batches of a specified size (imqConnectionFlowCount), ensuring periodic opportunities to deliver any accumulated control messages.
Connection flow control limits the number of payload messages (imqConnectionFlowLimit) that can be held pending on a connection, waiting to be consumed. When the limit is reached, delivery of payload messages to the connection is suspended until the number of messages awaiting consumption falls below the limit. Use of this feature is controlled by a boolean flag (imqConnectionFlowLimitEnabled).
Consumer flow control limits the number of payload messages (imqConsumerFlowLimit) that can be held pending for any single consumer, waiting to be consumed. (This limit can also be specified as a property of a specific queue destination, consumerFlowLimit.) When the limit is reached, delivery of payload messages to the consumer is suspended until the number of messages awaiting consumption, as a percentage of imqConsumerFlowLimit, falls below the limit specified by the imqConsumerFlowThreshold attribute. This helps improve load balancing among multiple consumers by preventing any one consumer from starving others on the same connection.
The use of any of these flow control techniques entails a trade-off between reliability and throughput; see Client Runtime Message Flow Adjustments for further discussion.
Table 18–6 lists connection factory attributes affecting client queue browsing and server sessions. The imqQueueBrowserMaxMessagesPerRetrieve attribute specifies the maximum number of messages to retrieve at one time when browsing the contents of a queue destination; imqQueueBrowserRetrieveTimeout gives the maximum waiting time for retrieving them. (Note that imqQueueBrowserMaxMessagesPerRetrieve does not affect the total number of messages browsed, only the way they are batched for delivery to the client runtime: fewer but larger batches or more but smaller ones. Changing the attribute's value may affect performance, but will not affect the total amount of data retrieved; the client application will always receive all messages in the queue.) The boolean attribute imqLoadMaxToServerSession governs the behavior of connection consumers in an application server session: if the value of this attribute is true, the client will load up to the maximum number of messages into a server session; if false, it will load only a single message at a time.
The Java Message Service Specification defines certain standard message properties, which JMS providers (such as Message Queue) may optionally choose to support. By convention, the names of all such standard properties begin with the letters JMSX. The connection factory attributes listed in Table 18–7 control whether the Message Queue client runtime sets certain of these standard properties. For produced messages, these include the following:
Identity of the user sending the message
Identity of the application sending the message
Transaction identifier of the transaction within which the message was produced
For consumed messages, they include
Transaction identifier of the transaction within which the message was consumed
Time the message was delivered to the consumer
You can use the connection factory attributes listed in Table 18–8 to override the values set by a client for certain JMS message header fields. The settings you specify will be used for all messages produced by connections obtained from that connection factory. Header fields that you can override in this way are
Delivery mode (persistent or nonpersistent)
Expiration time
Priority level
There are two attributes for each of these fields: one boolean, to control whether the field can be overridden, and another to specify its value. For instance, the attributes for setting the priority level are imqOverrideJMSPriority and imqJMSPriority. There is also an additional attribute, imqOverrideJMSHeadersToTemporaryDestinations, that controls whether override values apply to temporary destinations.
Because overriding message headers may interfere with the needs of specific applications, these attributes should only be used in consultation with an application’s designers or users.
The destination administered object that identifies a physical queue or topic destination has only two attributes, listed in Table 18–9. The important one is imqDestinationName, which gives the name of the physical destination that this administered object represents; this is the name that was specified with the -n option to the imqcmd create dst command that created the physical destination. (Note that there is not necessarily a one-to-one relationship between destination administered objects and the physical destinations they represent: a single physical destination can be referenced by more than one administered object, or by none at all.) There is also an optional descriptive string, imqDestinationDescription, which you can use to help identify the destination object and distinguish it from others you may have created.