Configuration information can be managed through procedures in the DBMS_AQADM package.

Initially, only SYS and SYSTEM have execution privilege for the procedures in DBMS_AQADM and DBMS_AQ. Users who have been granted EXECUTE rights to these two packages are able to create, manage, and use queues in their own schemas. The MANAGE_ANY AQ system privilege is used to create and manage queues in other schemas.

Users of the Java Message Service (JMS) API need EXECUTE privileges on DBMS_AQJMS and DBMS_AQIN.

Topics:

• Administrator Role

• User Role

• Access to Oracle Database Advanced Queuing Object Types

Oracle Database Advanced Queuing administrators of Oracle Database can create queues. When you create queues, the default value of the compatible parameter in DBMS_AQADM.CREATE_QUEUE_TABLE is that of the compatible parameter.

To enqueue or dequeue, users need EXECUTE rights on DBMS_AQ and either enqueue or dequeue privileges on target queues, or ENQUEUE_ANY/DEQUEUE_ANY system privileges.

You can grant or revoke privileges at the object level on queues. You can also grant or revoke various system-level privileges.

Table 4-1 lists all common Oracle Database Advanced Queuing operations and the privileges needed to perform these operations.

For an Oracle Call Interface (OCI) application to access a queue, the session user must be granted either the object privilege of the queue he intends to access or the ENQUEUE ANY QUEUE or DEQUEUE ANY QUEUE system privileges.

The EXECUTE right of DBMS_AQ is not checked against the session user's rights.

Oracle Database Advanced Queuing propagates messages through database links.

The propagation driver dequeues from the source queue as owner of the source queue; hence, no explicit access rights need be granted on the source queue. At the destination, the login user in the database link should either be granted ENQUEUE ANY QUEUE privilege or be granted the right to enqueue to the destination queue. However, if the login user in the database link also owns the queue tables at the destination, then no explicit Oracle Database Advanced Queuing privileges must be granted.

The export of queues entails the export of the underlying queue tables and related dictionary tables. Export of queues can also be accomplished at queue-table granularity.

Similar to exporting queues, importing queues entails importing the underlying queue tables and related dictionary data. After the queue table data is imported, the import utility executes the PL/SQL anonymous blocks in the dump file to write the metadata to the data dictionary.

The Data Pump replace and skip modes are supported for queue tables.

In the replace mode an existing queue table is dropped and replaced by the new queue table from the export dump file. In the skip mode, a queue table that already exists is not imported.

The truncate and append modes are not supported for queue tables. The behavior in this case is the same as the replace mode.

You cannot have more than 1,000 local subscribers for each queue.

Also, only 32 remote subscribers are allowed for each remote destination database.

Oracle Database Advanced Queuing does not support data manipulation language (DML) operations on queue tables or associated index-organized tables (IOTs), if any.

The only supported means of modifying queue tables is through the supplied APIs. Queue tables and IOTs can become inconsistent and therefore effectively ruined, if data manipulation language (DML) operations are performed on them.

Oracle Database Advanced Queuing does not support propagation from object queues that have REF attributes in the payload.

You cannot construct a message payload using a VARRAY that is not itself contained within an object.

You also cannot currently use a NESTED Table even as an embedded object within a message payload. However, you can create an object type that contains one or more VARRAYs, and create a queue table that is founded on this object type, as shown in Example 4-1.

CREATE TYPE number_varray AS VARRAY(32) OF NUMBER; 
CREATE TYPE embedded_varray AS OBJECT (col1 number_varray);
EXECUTE DBMS_AQADM.CREATE_QUEUE_TABLE( 
  queue_table           =>     'QT', 
  queue_payload_type    =>     'embedded_varray'); 

No Oracle Database Advanced Queuing PL/SQL calls resolve synonyms on queues and queue tables.

Although you can create synonyms, you should not apply them to the Oracle Database Advanced Queuing interface.

If you have created synonyms on object types, you cannot use them in DBMS_AQADM.CREATE_QUEUE_TABLE. Error ORA-24015 results.

Oracle Database Advanced Queuing currently does not support tablespace point-in-time recovery.

Creating a queue table in a tablespace disables that particular tablespace for point-in-time recovery. Oracle Database Advanced Queuing does support regular point-in-time recovery.

You can use Oracle Database Advanced Queuing with Virtual Private Database by specifying a security policy with Oracle Database Advanced Queuing queue tables.

While dequeuing, use the dequeue condition (deq_cond) or the correlation identifier for the policy to be applied. You can use "1=1" as the dequeue condition. If you do not use a dequeue condition or correlation ID, then the dequeue results in an error.

Propagation jobs are owned by SYS, but the propagation occurs in the security context of the queue table owner.

Previously propagation jobs were owned by the user scheduling propagation, and propagation occurred in the security context of the user setting up the propagation schedule. The queue table owner must be granted EXECUTE privileges on the DBMS_AQADM package. Otherwise, the Oracle Database snapshot processes do not propagate and generate trace files with the error identifier SYS.DBMS_AQADM not defined. Private database links owned by the queue table owner can be used for propagation. The user name specified in the connection string must have EXECUTE access on the DBMS_AQ and DBMS_AQADM packages on the remote database.

Propagation from object queues with BFILE objects is supported.

To be able to propagate object queues with BFILE objects, the source queue owner must have read privileges on the directory object corresponding to the directory in which the BFILE is stored. The database link user must have write privileges on the directory object corresponding to the directory of the BFILE at the destination database.

AQ propagation does not support non-final types. Propagation of BFILE objects from object queues without specifying a database link is not supported.

AQ propagation jobs are run by the Oracle Scheduler. Propagation may be scheduled in these ways.

• A dedicated schedule in which the propagation runs forever or for a specified duration. This mode provides the lowest propagation latencies.

• A periodic schedule in which the propagation runs periodically for a specified interval. This may be used when propagation can be run in a batched mode.

• An event based system in which propagation is started when there are messages to be propagated. This mode makes more efficient use of available resources, while still providing a fast response time.

The administrator may choose a schedule that best meets the application performance requirements.

Oracle Scheduler will start the required number of job queue processes for the propagation schedules. Since the scheduler optimizes for throughput, if the system is heavily loaded, it may not run some propagation jobs. The resource manager may be used to have better control over the scheduling decisions. In particular, associating propagation jobs with different resource groups can allow for fairness in scheduling which may be important in heavy load situations.

In setting the number of JOB_QUEUE_PROCESSES, DBAs should be aware that this number is determined by the number of queues from which the messages must be propagated and the number of destinations (rather than queues) to which messages must be propagated.

A scheduling algorithm handles propagation. The algorithm optimizes available job queue processes and minimizes the time it takes for a message to show up at a destination after it has been enqueued into the source queue, thereby providing near-OLTP action. The algorithm can handle an unlimited number of schedules and various types of failures. While propagation tries to make the optimal use of the available job queue processes, the number of job queue processes to be started also depends on the existence of jobs unrelated to propagation, such as replication jobs. Hence, it is important to use the following guidelines to get the best results from the scheduling algorithm.

The scheduling algorithm uses the job queue processes as follows (for this discussion, an active schedule is one that has a valid current window):

• If the number of active schedules is fewer than half the number of job queue processes, then the number of job queue processes acquired corresponds to the number of active schedules.

• If the number of active schedules is more than half the number of job queue processes, after acquiring half the number of job queue processes, then multiple active schedules are assigned to an acquired job queue process.

• If the system is overloaded (all schedules are busy propagating), depending on availability, then additional job queue processes are acquired up to one fewer than the total number of job queue processes.

• If none of the active schedules handled by a process has messages to be propagated, then that job queue process is released.

• The algorithm performs automatic load balancing by transferring schedules from a heavily loaded process to a lightly load process such that no process is excessively loaded.

The scheduling algorithm has robust support for handling failures. These are the common failures that prevent message propagation.

• Database link failed

• Remote database is not available

• Remote queue does not exist

• Remote queue was not started

• Security violation while trying to enqueue messages into remote queue

Under all these circumstances the appropriate error messages are reported in the DBA_QUEUE_SCHEDULES view.

When an error occurs in a schedule, propagation of messages in that schedule is attempted again after a retry period that is a function of the number of failures. After the retries have exceeded a system defined maximum, the schedule is disabled.

If the problem causing the error is fixed and the schedule is enabled, then the error fields that indicate the last error date, time, and message continue to show the error information. These fields are reset only when messages are successfully propagated in that schedule.