5 Tracking Business Flow Instances

This chapter describes how to track the status of business flow instances, including specifying and saving business flow search criteria, deleting and terminating instances, viewing the current state of an instance, recovering and aborting faults in an instance, viewing composite sensor values in an instance, viewing the initiating and participating composites in an instance, viewing resequencing groups in an instance, and viewing business flows that extend across domains.

This chapter includes the following section:

For more information about business flow instances and differences between instances in Release 12c and Release 11g, see Introduction to Business Flow Instances.

For information about managing faults on the Error Hospital page, see Recovering From Faults in the Error Hospital.

5.1 Tracking Business Flow Instances at the SOA Infrastructure or Partition Level

You can track the instances of business flows from the Flow Instances page of the SOA Infrastructure, individual partition, or individual SOA composite application. A business flow instance corresponds to an end-to-end business transaction. Business flows consist of a single SOA composite application or multiple SOA composite applications connected together to fulfill a specific business process.

Tracking business flow instances enables you to identify the following:

  • Determine what happened to a specific request

    • Has the request been received?

    • Has request processing completed or is it still in progress?

    • If processing has finished, has it completed successfully or did errors occur?

    • If processing is still in progress, is it stuck on an external request, awaiting a scheduled recovery, or in need of any administrative action?

  • Analyze and resolve a large number of failures

    For example, a runtime failure of a back end service (such as a database) has occurred. In high volume systems, a very large number of faults can occur in a short duration because of a single cause. These faults may be spread across SOA composite applications.You can identify the following under these circumstances:

    • Determine the root cause of the failure (typically done by analyzing one or two individual instances).

    • Find the number of requests with the same root cause of failure across different SOA composite applications.

    • Apply a bulk recovery/termination action to all instances with the same root cause of failure and recovery status (for example, a BPEL recovery or EDN recovery).

    • Specify a batch size.

    • Optionally perform a quick purge to remove the faults that have been handled or abort faults so that the next batch can be addressed.

You can manage business flow instances from the Flow Instances and Error Hospital pages at the following levels:

  • SOA Infrastructure level (provides access to all partitions)

  • Individual partition level

Note:

You can also manage business flow instances from the Flow Instances page of an individual SOA composite application. Flows that are initiated by a particular composite or passing through a composite are listed at the composite level. There are separate filter criteria for each. There is no Error Hospital page at the individual SOA composite application level. For information, see Tracking Business Flow Instances at the SOA Composite Application Level.

You can configure your environment to provide some administrators with access to the top level SOA Infrastructure Flow Instances and Error Hospital pages and other administrators with access to only the individual partition's Flow Instances and Error Hospital pages. For more information, see Managing SOA Folders and Work Manager Groups.

Access the Flow Instances page through one of the following options:

To access business flow instances in all partitions:

From the SOA Infrastructure Menu... From the SOA Folder in the Navigator... From the SOA Composite Menu...
  1. Select Home > Flow Instances.

  1. Expand the SOA folder.

  2. Click soa-infra.

  3. Click the Flow Instances tab.

  1. Select SOA Infrastructure.

  2. Click the Flow Instances tab.

To access business flow instances in an individual partition:

From the SOA Infrastructure Menu... From the SOA Folder in the Navigator...
  1. Select Manage Partitions.

  2. In the SOA Partitions column, click a specific partition.

  3. Click the Flow Instances tab.

  1. Expand SOA > soa-infra.

  2. Click a specific partition.

  3. Click the Flow Instances tab.

To access business flow instances for an individual SOA composite application, see Tracking Business Flow Instances at the SOA Composite Application Level.

The Flow Instances page displays the following details:

  • A utility for specifying and saving comprehensive instance and fault search criteria and clicking Search.

    Note:

    When you initially access the Flow Instances page, the Search Results table is empty. You must click Search to populate this table with business flow instance details.

  • A Search Results table that displays the flow ID that uniquely identifies a business flow, the composite initiating the business flow (business flows can span multiple composites), the state of the flow instance (for example, completed successfully, running, failed, recovery required, and so on), the instance start time, the last update to the instance, the partition in which this flow was initiated, the flow instance name (if set during design time in a BPEL process or Oracle Mediator service component), and a link to the log files. To display additional columns in the table, select View > Columns.

  • An Error Hospital button above the Search Results table that takes you to the Error Hospital page for a view of the results of the search criteria you executed on the Flow Instances page. This action does not require you to select any rows in the Search Results table. Use this option to view the error report for the flow instances currently displayed in the Flow Instances page. In the Error Hospital page, you can view aggregated fault statistics for these instances. You can also perform bulk recover and bulk abort operations on these flow instances without having to access each individual instance.

Note:

  • Business flow instances and faults data is delimited by the time period for which instances and faults are retrieved. The current delimiter is displayed to the right of the Search Results table title. The default value is 24 hours. You can change this value with the Default Query Duration property on the SOA Infrastructure Common Properties page. For information, see Configuring the Audit Trail, Payload Validation, and Default Query Duration.

  • When you undeploy and redeploy the same SOA composite application, you cannot retrieve the business flow instances created prior to composite undeployment. The business flows are still available. However, the active instances are aborted or they are untouched based on your configuration while redeploying. For more information, see Redeploying SOA Composite Applications.

You can perform the following business flow instance tasks:

5.1.1 Specifying and Saving Business Flow Search Criteria

The Search Options section enables you to specify and save comprehensive instance and fault search criteria. Search criteria results are displayed in the Search Results table. Saved search criteria are user-specific, and are not common across users. Each Oracle Enterprise Manager Fusion Middleware Control user can save their own preferred searches. Upon creation, saved searches are displayed on the Dashboard page and also as links on the Error Hospital and Flow Instances pages.

To specify and save business flow search criteria:

  1. Click the Search Options icon to display the Search Options section. The Search Options section displays in a sliding panel and may not be visible on the page at all times.
  2. See the following sections to specify and save comprehensive instance and fault search criteria.
5.1.1.1 Executing Predefined Business Flow Instance Searches

You can quickly find business flows without entering any search criteria by selecting a predefined search option. Results are displayed in the Search Results table. The searches are constrained by a predefined time period. The default time period is 24 hours. This value can be changed by modifying the Default Query Duration property in the SOA Infrastructure Common Properties page, accessible under SOA Administration in the SOA Infrastructure menu.

The following options are available:

  • Recent Instances: Displays recent instances.

  • Instances With Faults: Displays recent instances that have faults. This predefined search option is also available on the Error Hospital page, where you can select it from the Report Filters list.

  • Recoverable Instances: Displays recent instances awaiting recovery.

At the top of the Search Options section, select the option for which to search.

or:

The search results are displayed in the Search Results table (for this example, the search results for the Recoverable Instances option are displayed).

For more information about executing predefined searches, select Help > Help for This Page from the weblogic main menu on the Flow Instances page.

For information about saved searches, see Using the Search Options Toolbar.

5.1.1.2 Using the Search Options Toolbar

The Search Options toolbar enables you to perform search-related tasks, such as executing predefined or custom searches, saving search criteria, resetting displayed search criteria, bookmarking searches, and adding and removing filters. By default, only predefined searches can be invoked. You can extend the list of available searches by saving custom searches. The Search Options toolbar displays in a sliding panel and may not be visible on the page at all times. If not already open, you can invoke it by clicking the large Search Options icon.

Go the toolbar in the Search Options section.

The following options are available.

Element Description
Reset icon

Click to reset the search fields in the currently invoked saved search to the last saved values. This is useful when you have modified a saved search and want to restart the query building process.

Save icon

Click to save your current search criteria. This saves both the selected search fields and their values, enabling you to run the identical search at a later time and view a fresh set of results.

You must provide a name when saving a search. You cannot overwrite an existing saved search, but you can save it with a different name. You can delete the saved searches you created. To manage your saved searches, select All Saved Searches from the Search Options list.

Bookmark icon

Click to bookmark your current search criteria. A message is displayed with a URL containing the search parameters. Copy the URL to a browser bookmark window, email, or chat. The generated URL includes information about both the selected search fields and their values. This enables you to run the identical search at a later time and view a fresh set of results.

For more information about the Search Options toolbar, select Help > Help for This Page from the weblogic main menu on the Flow Instances page.

5.1.1.3 Adding and Removing Search Filters

You can add and remove the search filters that are displayed for configuration. Filters are the attributes of a business flow that you can use to create your search criteria. After selecting a filter, you can customize it in the Search Options section. All filters with specified search criteria are added together (treated as logical AND values) when the resulting query is run. Empty fields are ignored.

  1. Click Add/Remove Filters.

    The filters available for selection are displayed. Some filters are interdependent. When you remove them, your current search parameters may be reset. The time filter is required and cannot be removed.

  2. Select and deselect appropriate filters to use in search criteria, and click OK.
  3. Click Search.

    The selected criteria are displayed in the Search Options section. You can further customize the selected filters, as described in Configuring and Saving Business Flow Instance Search Filter Criteria.

5.1.1.4 Configuring and Saving Business Flow Instance Search Filter Criteria

You can configure parameters for each search filter selected in Adding and Removing Search Filters to create a business flow instances search query. Search results are displayed in the Search Results table.

  1. Go to the Search Options section.
  2. Configure parameters for appropriate search filters. Filters left blank are ignored. You do not need to remove them. For more information about configuring each filter, select Help > Help for This Page from the weblogic main menu on the Flow Instances page.
    Element Description

    Time

    Restrict your search query to a specific time in the past. A time filter is required to search for business flows. At least one time filter is required. If more than one time filter list is displayed, ensure that you specify values for both. Click Options to display the filters available for selection. Once selected, you can specify the time periods to be searched.

    • Instance Created (displays by default and cannot be deselected)

    • Instance Updated

    • Fault Occurred

    You can also specify a custom time period for which to search by selecting the Custom time period checkbox. This search is for all three time periods (Instance Created, Instance Updated and Fault Occurred).

    Composite

    Restrict your search query for business flows to a specific composite.

    • If searching at the partition level, only instances that were initiated and participated in by SOA composite applications in that partition are returned.

    • If searching at the SOA Infrastructure level, instances initiated or participated in by SOA composite application in any partition are returned.

    Perform the following tasks:

    • Select Initiating if you want to limit your search only to the business flows that started in the selected composite. To search for all business flows in that composite, select Participating.

    • Select the partition to search. If you access the Flow Instances page at the individual partition level, that partition is already selected and you cannot change it. If you do not select a partition, all are searched.

    • Select the specific SOA composite application name from the list or click Search to specify a complete or partial name for which to search. The partial name search supports only entering the beginning part of the name and is case sensitive. If you do not specify a composite, all are searched.

    Resequencer

    Select the resequencing groups for which to search. Use this filter to limit your search only to business flows in which a resequenced component participated. If you leave this section blank, this search filter is ignored. The resequencer in Oracle Mediator rearranges a stream of related but out-of-sequence messages into a sequential order.

    • Any Group: Select to search for all resequenced flows in all groups.

    • Specific Group: Select to find the business flow associated with a specific resequencing group. Specify the group's name and location. The location is the Oracle Mediator component and SOA composite application revision containing the group. The group name filter returns only one group instance.

    For more information about resequencing groups, see Monitoring Resequencing Groups.

    Flow Instance

    To find a specific business flow instance, identify the instance in at least one of the following ways:

    • Flow ID: Select to search by flow ID, initiating ECID, or correlation flow ID.

    • Instance Name: Specify the name or title of the flow instance that you created at design time in Oracle JDeveloper.

    • Composite Instance Name: Specify the name or title of the composite instance name. SOA composite applications upgraded from Release 11g can contain a composite instance name or title.

    • Add Sensor Value (Up to 6): Click to specify up to six composite sensor name and value pairs.

    Note: You can populate the Sensor Name fields by either manually entering names or by using the lookup feature. If you use the lookup feature, you must specify a composite for each sensor you add in the search. All sensor-based search conditions are connected with a logical AND. However, if you specify multiple sensors from different composites, you likely end up with an empty results table. As a best practice, stick with sensors from the same composite because they are more likely to have matching values in the same flow instance.

    Note these additional sensor guidelines:

    • All sensor conditions are connected with a logical AND. You cannot enter alternative sensor names. The search attempts to find an instance that matches all the names.

    • When manually entering names, you must provide the exact sensor name and value.

    • You must specify both the name and value for each sensor.

    • Leaving the Value field empty is not supported (no results are returned). It is not possible to use the search to find all instances in which a sensor is defined, regardless of the value.

    • If you want to use operators other than equals (=), use the lookup feature.

    • If you know the exact sensor name, but not the composite, use the main search, which attempts to find instances across all composites. However, if you want to see a list of sensors from which to select, use the lookup feature. You also must know the composite name, because the lookup feature makes you specify the composite first (it is looking for sensors one composite at a time). Also, all composites are listed: those with sensors defined for them and those without. Trying every composite is not efficient.

    State

    Restrict your search to only the business flows in a given state. For descriptions of states, see Viewing the Current State of the Business Flow Instance.

    Select Active to search active instances. Active instances have not reached one of the terminal states. The list is refreshed to display the following selections for further filtering:

    • All active

    • Recovery

    • Suspended

    • Running

    Select Inactive to search inactive instances. Inactive instances have reached one of the terminal states. The list is refreshed to display the following selections for further filtering:

    • All inactive

    • Completed

    • Failed

    • Aborted

    Fault

    Restrict the search for business flows to only those with faults. If you leave this field blank, this filter is ignored. Select to specify the types of faults for which to search.

    • All Faults: Select to search for business flows containing any type of fault.

    • Recovery Required: Select to search for stuck flows awaiting human recovery to proceed. To further specify a particular type of recovery, use the Fault Recovery Type filter.

    • Nonrecoverable: Select to search for flows containing nonrecoverable faults.

    • Recovered: Select to search for flows that contain at least one recovered fault.

    • System Auto Retries: Select to find the faulted flows in which system automatic retries occurred.

    To further customize fault filtering, click More next to the Faults filter to display additional filtering attributes.

    • Fault Recovery Type: Filter your search for faulted business flows to stuck flows awaiting a particular type of recovery action. This field is available when you select Recovery Required in the Fault filter and click More. If you leave this field blank, the Fault Recovery Type filter is ignored. The fault recovery types are as follows:

      Admin Recovery

      BPEL Activity Recovery

      BPEL Invoke Message Recovery

      BPEL Callback Message Recovery

      EDN Recovery

      Mediator Recovery

      Human Workflow Recovery

      Rejected Message Recovery

    • Fault Type: Filter your search for faulted business flows. If you leave this field blank, the Fault Type filter is ignored. Select one of the following to restrict your search to only the flows containing that fault type.

      System: Network errors or other types of errors such as a database server or a web service being unreachable.

      Business: Application-specific faults that are generated when there is a problem with the information being processed (for example, a social security number is not found in the database).

      OWSM: Errors on Oracle Web Service Manager (OWSM) policies attached to SOA composite applications, service components, or binding components. Policies apply security to the delivery of messages.

    • Fault Owner: Select the specific component, service, or reference in which the fault was handled (also known as the fault owner). Use this filter to further narrow down your search for faulted business flows. If you leave it blank, the Fault Owner filter is ignored. The fault owner is similar to the fault location, but they are not always the same. The fault owner is the component that executes the recovery or from where to attempt a recovery, in which case the flow is re-executed. The fault location is the component that created or caused the fault.

    • Fault Details: Filter a search for faulted business flows. If you leave all fields blank, the Fault Details filter is ignored. Specify at least one of the following fault details. To find only faults for which these values are not set, enter NOT SPECIFIED in the search field.

      Error Message Contains: Use to find only faulted business flows with the same error message text. You can enter any part of the message. This search is case sensitive.

      Fault Name: Use to find only faulted business flows with a specific descriptive fault name such as NegativeCredit. You must enter the exact name (the entire string). This search is case sensitive.

      Fault Code: Use to find only faulted business flows with the same fault code.

      To customize fault search criteria, click More next to Fault Details to display additional configuration fields such as HTTP host, JNDI name, event name, and event namespace.

    User

    Finds all instances associated with a specific Oracle Enterprise Manager user or business owner. The user is set only in SOA composite applications in which user credentials are explicitly propagated (for example, when a composite has a user authentication OWSM policy attached to the composite web service).

  3. If you want to save search criteria for future use, click the Save Search icon to specify a name.

    Your saved search is then available for selection in the Search Options dropdown list and in the Report Filters section of the Error Hospital page.

  4. When search criteria creation is complete, click Search.
  5. View search results in the Search Results table.
  6. Click Error Hospital to display aggregated fault details in the search results, including the fault name, total number of faults, faults requiring recovery, unrecoverable faults, recovered faults, and automatic fault retries. No selections are necessary, as this aggregates the fault information for all flow instances that are already displayed in the table.

5.1.2 Deleting or Terminating Business Flow Instances

You can delete or abort business flow instances in the Search Results table.

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.

    The Search Results table is populated with business flow instances.

  2. Select a specific business flow or select multiple business flows by pressing Shift-Click.
  3. Click the Actions list and select to delete or abort a business flow instance. You can also click the Deleted Selected or Abort Selected icons above the table.
    • Delete Selected: Deletes selected business flow instances in the table. When you select to confirm the deletion of instances, the page is refreshed and the deleted instances are removed from the list. You can no longer access the flow trace details of the instance once it is deleted.

    • Abort Selected: Terminates selected business flow instances in the table. When you confirm to abort the selected instances, the page is refreshed and the states of the aborted instances are displayed as Aborted. In some cases there may be a delay in updating the instances state. You still can access the flow trace and audit trail details of the instance after it has been aborted.

      To abort a large number of instances based on search criteria without selecting each one individually, do not use the Abort Selected action on the Flow Instances page. Instead, use the Bulk Abort option on the Error Hospital page.

Note:

If you selected a running instance, the Abort option is available for stopping a running instance.

5.1.3 Viewing the Current State of the Business Flow Instance

You can view the components that comprise the business flow instance. Business flows can consist of multiple SOA composite applications connected together in a flow.

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.

    The Search Results table is populated with business flow instances.

  2. In the Flow State column of the Search Results table, view the state of a specific business flow instance.

    The business flow instance can exist in one of the following states:

    State Description

    Running

    The instance is still running (for example, it may be stopped in a human workflow service component awaiting an approval). There are no recoverable faults associated with this state and the instance has not yet successfully completed.

    Recovery

    A fault in the instance is awaiting recovery. See Recovering from Faults in a Business Flow Instance.

    Suspended

    The instance is currently suspended. This is related to the migration of one version of the composite to another. This is applicable to Oracle BPM Suite components.

    Aborted

    The instance has been explicitly terminated by a user. When a composite is undeployed, running instances are marked aborted. Upon redeployment, you are given a choice to abort or keep running.

    Completed

    The instance successfully completed and there are no recoverable and nonrecoverable faults.

    Failed

    The instance completed, but there are nonrecoverable faults.

  3. In the Flow ID column, click a specific flow ID number to access more specific details about the state of the business flow instance.

    The Flow Trace page is displayed and shows the path taken by the flow (entry points, components executed, and exit points). The flow trace is available only if the Audit Level property is set to Production or Development. If the audit level is set to Off, the trace view is replaced by a component instance listing. You can recover from faults using the Faults tab on the Flow Trace page.

  4. In the Trace table, click a specific service component to view the audit trail and process flow. You can also view XML of the flow trace by selecting the Actions list.

Note:

If you invoke an Oracle SOA Suite Release 12c direct binding service from Oracle Service Bus Release 11g (releases 11.1.1.4 through 11.1.1.7), no entry is displayed in the Trace table of the Flow Trace page in Oracle Enterprise Manager Fusion Middleware Control Release 12c that shows the direct binding service getting invoked from Oracle Service Bus. The display of this type of invocation is not supported.

In Oracle SOA Suite Release 12c, direct binding service invocation from Oracle Service Bus Release 12c is displayed in the Trace table of the Flow Trace page in Oracle Enterprise Manager Fusion Middleware Control.

A direct binding service is also displayed in a flow with Oracle Enterprise Scheduler or Oracle Service Bus at the start or middle of the flow.

5.1.4 Recovering from Faults in a Business Flow Instance

You can recover from faults identified as recoverable in the selected business flow.

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.

    The Search Results table is populated with business flow instances.

  2. In the Search Results table, identify instances with faults labeled as Recovery in the Flow State column. Faults in this state are awaiting recovery.

  3. Select the business flow instance, and click Show Details.

    The Faults tab is displayed below the Search Results table. This tab displays the error message, the fault owner (the service component, service binding component, or reference binding component in which the fault was handled), the composite in which the fault occurred, the time of the fault, if the fault is recoverable, and links to the fault logs. To display additional columns, select View > Columns. Select Reset Columns to remove all customizations such reordering columns and showing and hiding columns on the Faults table.

  4. In the Error Message column, place your cursor over the error message to display the entire error message or click the red fault icon to display a popup dialog from which to copy the message. You can also place your cursor over the link or icon in the Recovery column to see what type of recovery is required.

  5. Select a method for performing fault recovery. Either method enables you to automatically perform a recovery or perform a recovery through human intervention. This human intervention action is defined with the ora-human-intervention action in the fault-policies.xml file that you designed in Oracle JDeveloper.

    To Perform Fault Recovery From... Go To Step...

    The Recover list above the table

    6

    The Recovery Required message in the Recovery column

    7

  6. Select the row in the table on which to perform fault recovery.

    1. From the Recover list above the table, select a fault recovery action to perform. The Retry and Abort options are available for all recoverable faults. The Continue, Rethrow, and Replay options depend upon the contents of the fault policy file attached to the SOA composite application.

      The Fault Details and Recovery Options dialog is displayed.

    2. Go to Step 8.

  7. In the Recovery column, click the Recovery Required message.

    A dialog describing the fault is displayed.

    1. Click Retry or Abort to automatically retry or abort the fault or click More Recovery Options to perform human intervention.

      If automatic recovery is successful, the flow in the Search Results table of the Flow Instances page is refreshed to display a different flow state (for example, recovery is changed to completed). In some cases, there may be a delay in getting the updated state and consequently the display does not show the changed state. However, in all cases, a confirmation message is displayed that indicates that the recovery request was sent. Perform an explicit page level refresh either on the button at the fault level or the Refresh button at the page level.

      If you clicked More Recovery Options, the Fault Details and Recovery Options dialog is displayed.

    2. Go to Step 8.

  8. In the Recovery Options section of the Fault Details and Recovery Options dialog, view the options for performing fault recovery. Other recovery types such as BPEL message recovery show different sets of options. Any type of fault can be recovered from this console, such as fault policy-triggered recovery and BPEL invoke, callback, and activity message recovery. As with Release 11g, you can also browse to the BPEL process service engine Recovery tab and perform invoke, callback, and activity recovery.

  9. From the Recovery Action list, select an action to perform.

    Action Description Action is Available for...

    Abort

    Terminates the entire instance.

    BPEL process and Oracle Mediator

    Retry

    Retries the instance directly. An example of a scenario in which to use this recovery action is when the fault occurred because the service provider was not reachable due to a network error. The network error is now resolved.

    BPEL process and Oracle Mediator

    Continue

    Ignores the fault and continues processing (marks the faulted activity as a success).

    BPEL process

    Rethrow

    Rethrows the current fault. BPEL fault handlers (catch branches) are used to handle the fault. By default, all exceptions are caught by the fault management framework unless an explicit rethrow fault policy is provided.

    BPEL process

    Replay

    Replays the entire scope again in which the fault occurred.

    BPEL process

  10. From the Variable list, select a variable. The content of this variable is displayed in the Value field. For this example, the variable crInput is selected. This variable is used in an invoke activity and contains an incorrect social security number value.

  11. Enter the correct value in the Value field. For this example, the social security number is edited to begin with 1.

  12. Click Recover.

For information about designing a fault policy, see Section "Handling Faults with the Fault Management Framework" of Developing SOA Applications with Oracle SOA Suite.

5.1.5 Viewing Composite Sensor Values in a Business Flow Instance

The Composite Sensor Values tab displays the values of composite sensors detected in the selected business flow. Composite sensors provide a method for implementing trackable fields on messages. Composite sensors may be included in the service and reference binding components of the SOA composite application associated with the flow. This tab only has data displayed if the sensors are defined for the composite and the values captured by the selected business flow are not null.

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.

    The Search Results table is populated with business flow instances.

  2. In the Search Results table, select a specific business flow instance, and click Show Details.
  3. Click the Composite Sensor Values tab.

    This tab displays composite sensor names and their values included in this flow.

    For more information about composite sensors, see Chapter "Defining Composite Sensors" of Developing SOA Applications with Oracle SOA Suite.

5.1.6 Viewing the Initiating and Participating SOA Composite Applications in a Business Flow Instance

You can view all the SOA composite applications participating in a business flow instance. This tab displays a list of the composites that are associated with this flow, their sequence in the flow (the composite that initiated the business flow and all subsequent composites that participated in the business flow), the time at which a flow entered the specific composite, and links to logs.

To view the initiating and participating composites in a business flow instance:

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.
  2. In the Search Results table, select a specific business flow instance, and click Show Details.
  3. Click the Composites tab.

    The initiating and participating SOA composite applications included in this business flow instance are displayed. The Name column shows the value of the flow instance title (In Release 11g, you specified the title or name for each composite instance). In Release 12c, this value is displayed under the Name column.

5.1.7 Viewing Resequencing Groups in a Business Flow Instance

The Resequencing Groups tab is only displayed if a resequenced component participated in the flow. Such a flow instance is indicated by a small triple-arrow icon. The resequencer in Oracle Mediator rearranges a stream of related but out-of-sequence messages into a sequential order. The groups are ordered by the last processing time (most recent first). Alternatively, you can reorder them by the group name or state.

You cannot select the rows in the table or perform any actions under the Resequencing Groups tab on the Flow Instances page.

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.
  2. In the Search Results table, select a specific business flow instance that includes a resequencing group, and click Show Details.
  3. Click the Resequencing Groups tab.

    The group name, group location, group state, and last processed time are displayed.

  4. In the Group column, click the name to display group details.

    If a resequencing group has faults, you can view the faults associated with a resequenced flow instance by clicking the Faults tab. Recoverable faults can be recovered from the Faults tab.

  5. Click All Resequencing Groups to access the Mediator Resequencing Groups page to view details about all defined groups, such as the group name, the Oracle Mediator and SOA composite application in which the group is included, the group state (for example, timed out or faulted), the last processing time, and the total number of backlogged messages.

For more information, see Monitoring Resequencing Groups.

5.1.8 Viewing Business Flows that Extend Across Domains

A flow correlation ID is a GUID value that uniquely identifies a business flow that extends across domains. The flow correlation ID differs from the flow ID, which uniquely identifies a business flow within an Oracle SOA Suite domain. A business flow that crosses two Oracle SOA Suite domains has a single flow correlation ID and two flow IDs. Similarly, a business flow that originates outside of Oracle SOA Suite cannot create a flow ID because it does not have access to the Oracle SOA Suite schema. However, it can create a flow correlation ID.

To view business flows that extend across domains:

  1. Specify search criteria in the Search Options section as described in Configuring and Saving Business Flow Instance Search Filter Criteria, and click Search.
  2. In the Search Results table, select Columns > Correlation Flow ID from the View main menu.

    Any flow correlation IDs are displayed in the table.

    When a SOA composite application submits an Oracle Enterprise Scheduler request, the flow correlation ID, and not the flow ID, is propagated to Oracle Enterprise Scheduler and displayed in the Submission FlowID field of the Execution Status section of the Request Details page. The Oracle Enterprise Scheduler submitter flow ID associated with a request corresponds to the Dynamic Monitoring Service (DMS) flow ID. The DMS Flow ID corresponds to the SOA flow correlation ID.

5.1.9 Partition-Level and Composite-Level Search Results for Redeployed Composites with the Same Revision Number

A partition-specific search shows all versions of a SOA composite application (even those that were redeployed with the same revision number), whereas a composite-specific search shows only the versions of the current deployment. Because the revision numbers do not change, it is not possible to determine which row in the Search Results table of the Flow Instances page belongs to which version. This can be confusing when it appears in the partition-level search, but not in the composite-level search. To prevent this situation, increment the revision number between deployments. This makes multiple deployments more evident.

For example, assume you perform the following steps:

  1. Deploy composite A and create several instances. Both a composite-specific search and a partition-specific search show these instances.
  2. Redeploy and overwrite composite A and create instances. The composite-specific search shows instances from the first revision in addition to the latest version. A partition-specific search shows instances from the first revision in addition to the latest version.
  3. Undeploy and redeploy composite A and create instances. The composite-specific search only shows instances from the latest version. The partition-specific search shows all instances from the first and second versions in addition to the existing version.