book
Article ID: KB0102289
calendar_today
Updated On:
Description
The hotfix can be downloaded from the TIBCO Support Customer Portal Web UI,
using your username and password for the TIBCO Support Web.
Once logged on the hotfix can be found under the Downloads Menu:
AvailableDownloads/AMX_BPM/4.1.0/TIB_amx-bpm_4.1.0_hotfix006
Listed below and in the Resolution section is a summary of updates included.
Refer to the associated readme document for any additional information.
================================================================================
Closed Issues in 4.1.0_HF-006 (This Release)
Note: Where the closed issue is related to a TIBCO Support case, the relevant
case number (CN-xxxxxxxx) or Support Request ID (SR-xxxxxx) is given in brackets
after the Issue number.
WRM-10126 (CN-01539679)
-----------------------
If a service process halted as a result of reaching its
MAX_TASKS_PER_INSTANCE_REACHED threshold value (see WRM-9282 below), the halt
did not propagate back to the parent process.
WRM-10123 (CN-01521867)
-----------------------
When a sub-process returned case data to its parent, if the data contained an
attribute that was not present in the case data model used by the parent, the
parent process halted.
Now, if this happens:
* The unknown attribute is simply removed, allowing the parent process to
continue.
* A WARN message is written to the BPM log file. For example:
[WARN] - {BX_UNKNOWN_XML_ATTRIBUTES_REMOVED} - Process[Root Task] Instance
[pvm:0a125]: removed unknown attributes from XML value...
* A BX_UNKNOWN_XML_ATTRIBUTES_REMOVED message is displayed in the audit trail.
NOTE: The audit message is only displayed if you add the
BX_UNKNOWN_XML_ATTRIBUTES_REMOVED message identifier to the ec-probe-rules.xml
file. To do this:
1. Make a backup copy of the <CONFIG_HOME>\bpm\<bpm_app_name>\configuration\
ec-probe-rules.xml file.
2. Open the file in a text editor and find the <messageIds> section of the
AuditMessages rule. For example:
<rule name="AuditMessages" ruleOrder="800">
<filter>
<severities>
<severity>AUDIT</severity>
</severities>
<messageIds>
<messageId>BX_INSTANCE_TASKS_CREATED</messageId>
<messageId>BX_INSTANCE_TASKS_COMPLETED</messageId>
...
</messageIds>
3. Add the following line to the <messageIds> section:
<messageId>BX_UNKNOWN_XML_ATTRIBUTES_REMOVED</messageId>
4. Save and close the file.
The BPM runtime monitors this file dynamically, so the change takes effect as
soon as the file is next checked. You do not need to restart any component of
the ActiveMatrix BPM system.
WRM-10121 (CN-01540729)
-----------------------
Process instance execution and work item scheduling performance problems were
encountered when the system was handling a high rate of incoming cases. For
example, there was a significant delay between completing one user task and
starting the next one.
The problem occurred when processing global signal event handlers.
(Specifically, adding a new waiting receiver to the signal, or subsequently
removing that receiver, caused a processing bottleneck where all threads within
all nodes blocked on the same signal definition.)
WRM-10113
---------
When purging audit data, data for sub-process instances could be incorrectly
purged when the parent process instance had not yet completed, resulting in an
incomplete audit trail. This error could occur if the ActiveMatrix BPM 4.1
system had been upgraded from a pre-4.0 version, when purging audit data that
had been migrated from the pre-4.0 version.
Updated migration scripts are provided in the <TIBCO_HOME>\bpm\4.1\scripts\
migration\database folder. If you are upgrading to ActiveMatrix BPM 4.1.0_HF-006
from a pre-4.0 version, you should use these scripts when you perform the
post-upgrade task to migrate existing audit data so that it is still available
after the upgrade. See "Migrating Data Stored in Event Collector", in the TIBCO
ActiveMatrix BPM 4.1 Installation and Configuration guide, for detailed
information about this task, but note that the task now requires an additional
step, like this:
1. Run install-migration-tools.sql to install the stored procedures and create
the ec_migration table.
2. Run the amxbpm_migrate_events stored procedure, to migrate the audit data to
the new database tables.
3. *NEW* Run the amxbpm_set_lifecycle_ids stored procedure, introduced by
WRM-10113. This corrects the problem in the migrated database tables,
ensuring that when data is purged, sub-process instance data is only purged
when its parent process instance has already completed.
amxbpm_set_lifecycle_ids Stored Procedure Parameters
----------------------------------------------------
amxbpm_set_lifecycle_ids (batch_size, max_updates, insert_active,
set_orphaned_as_root)
where:
* batch_size - is the number of rows to update before performing a commit.
* max_updates - is the maximum number of updates to perform before exiting.
* insert_active - must be set to '1' when migrating to version 4.1.
* set_orphaned_as_root - defines how any orphaned sub-process data (which may
exist if only some audit data was migrated from the
old version) is treated when data is purged:
0 - Data for orphaned sub-process instances will not
be purged.
1 - Data for orphaned sub-process instances will be
purged.
4. Run remove-migration-tools.sql to remove the migration tools.
WRM-10084 (CN-01535994, CN-01487999)
------------------------------------
After deploying a new major version of a case data model, a StackOverFlowError
occurred when trying to open a work item.
WRM-10045 (CN-01535912)
-----------------------
In a client application, a call to the getProcessInstanceAudit operation (in the
EventCollectorReportService) returned an exception if the specified process
instance did not exist in the ec_event view. This could happen if either the
specified process instance did not exist, or if the audit data for that process
instance had been purged.
Now, the getProcessInstanceAudit operation returns a successful response that
contains no entries. For example:
<xml-fragment>
<endPosition>0</endPosition>
<startPosition>0</startPosition>
<totalItems>0</totalItems>
<processInstanceRef>
<instanceId>jnlkjn</instanceId>
</processInstanceRef>
</xml-fragment>
WRM-10023 (CN-01533640)
-----------------------
A BPM background job periodically deletes completed rows from the PVM_WORK_ITEM
table. If Process Manager tried to update a row in the PVM_WORK_ITEM table at
the same time as the background job tried to delete it, the following error
occurred (when using an Oracle database):
ORA-00060: deadlock detected while waiting for resource
With this fix, the Process Manager update now ignores already completed rows.
(See also WRM-9982.)
WRM-10017 (CN-01532586)
-----------------------
Additional WARN level logging has been added to identify any errors that occur
when a BPM node attempts to compile a process definition. DEBUG level logging
has also been enhanced to provide additional detail about the cause of any
such errors.
WRM-10010 (CN-01458884)
----------------------
If a BPM node fails (or is otherwise deemed to be inactive), a periodic recovery
thread attempts to recover any instances related to that node. An error in the
recovery processing sometimes caused a null pointer exception. This resulted in
the rest of the recovery process repeatedly failing to complete.
WRM-9983 (CN-01531013)
----------------------
When an organization model was upgraded and privileges were removed, the
privileges were only removed from fixed organization units, not from dynamic
organization units. With this fix, the privileges are now also removed from the
existing dynamic organization unit and position instances when the extension
points are next processed.
WRM-9982 (CN-01529439, CN-01521867)
-----------------------------------
When Process Manager processes a row in the PVM_WORK_ITEM or PVM_REQUEST_QUEUE
database tables, it sets that row's STATUS value to 2 (DONE). A background job
periodically cleans these rows up. However, if the number of such rows grew too
large, performance problems were noticed when the background job was processing
work, particularly if Process Manager was configured to use large batch sizes
(using com.tibco.bx.groupRQ.batchSize).
With this fix, Process Manager now deletes a row in the PVM_WORK_ITEM or
PVM_REQUEST_QUEUE database tables when it has processed it, instead of setting
it to DONE. The following new JVM properties are used to enable this as the
default behavior:
com.tibco.pvm.deleteWorkItem=true
com.tibco.pvm.deleteRequestQueueItem=true
If you want to continue to use the original behavior instead, update the JVM
configuration for the ActiveMatrix BPM nodes (using ActiveMatrix Administrator)
to set these properties to false:
com.tibco.pvm.deleteWorkItem=false
com.tibco.pvm.deleteRequestQueueItem=false
then restart the BPM node.
WRM-9981 (CN-01529123)
----------------------
Attempted retries of halted process instances sometimes failed with the
following exception in the BPM log file:
[ERROR] com.tibco.bx.core.BxSystemImpl - [ERROR] - {BX_TASK_REPORTED_ERROR}
- BX-800009: Task reported an error ... com.tibco.bx.core.faults.BxException:
... Caused by: java.lang.IndexOutOfBoundsException:
WRM-9963 (CN-01528379)
----------------------
Service process calls were taking a long time to respond to the main process.
This problem occurred if the PVM_REQUEST_QUEUE database table contained rows for
process applications (modules) that were not running. This could happen if, for
example, large numbers of process instances were cancelled when the process
application itself was not running.
With this fix, performance improvements have been made in this area. Additional
WARN messages are also now written to the BPM log file to identify messages that
cannot be processed because a module is either not running or not available.
For example:
RequestQueueItem[qid:608, id:14805]: module[22] is not ready, rescheduling
event with lower priority
RequestQueueItem[qid:608, id:14805]: module[22] is not ready, rescheduling
event for job[103]
You can then move or delete the affected messages or fix the problem with the
indicated module.
WRM-9949 (CN-01524734)
----------------------
In Workspace, when viewing the details of a process instance (in the Process
Instance Data dialog), unexpected custom attributes were displayed. These were
attributes prefixed with "v_", for example "v_b6ab020b.parameters" or
"v_48fe6ccc.annualSubReq", which had not been defined as data fields in the
process when it was defined.
These fields were internal fields (generated by Process Engine) that should not
have been displayed. These fields were returned by the queryProcessInstances or
queryProcessInstancesAlt operations on the ProcessManagerService service of the
ActiveMatrix BPM public web service API. With this fix, these fields are
filtered out and not displayed, either in Workspace or in a client application
that uses the queryProcessInstances or queryProcessInstancesAlt operations.
NOTE: Any custom fields in a process prefixed with either "v_" or "BX_" will
also be filtered out and not displayed.
WRM-9877 (CN-00708438, CN-00728410, CN-00712381, CN-00800718)
-------------------------------------------------------------
Setting TRACE level logging for the com.tibco.pvm logger sometimes resulted in
an exception that caused a process instance to halt. The following exception was
reported to the BPM log file:
com.tibco.pvm.api.exceptions.PmModelException: PVM-MODEL-102001: Invalid
Object:
WRM-9840 (CN-01519185)
----------------------
Following migration, process instances were hanging (using CPU). This occurred
because migration rules had been specified where the "from" version was the same
as the "to" version.
WRM-9817 (CN-01458313)
----------------------
Security enhancement: Avoid disclosing details about authentication failures
under certain circumstances.
WRM-9800 (CN-01484928, CN-01488930, CN-01523498, CN-01536193, CN-01541007)
-------------------------------------------------------------------------
When using an ActiveMatrix BPM Developer Server, errors like this were written
to the BPM log file:.
[ERROR] com.tibco.pvm.system.PvmSystem - TxBlock: Error Running Block -
Last State: STARTED_TRANSACTION com.tibco.pvm.db.common.dao.util.fault.
IpmPersistenceFault:
PVM-PERSIST-100001: Database Query Error: Engine.lockEngineById
.
.
Caused by: org.hibernate.MappingException: Named query not known:
Engine.lockEngineById
This error occurred because the 'Engine.lockEngineById' named SQL query, was
missing from the Postgres database implementation used by the ActiveMatrix BPM
Developer Server.
NOTE: This issue did not affect ActiveMatrix BPM systems using Oracle, SQL
Server or DB2 databases.
Resolution
WRM-9789 (CN-01469457)
----------------------
Upgrading an already deployed Business Data project to a new minor version
sometimes failed if:
* the project depended on another Business Data project, and
* both projects had been initially deployed on an earlier version of
ActiveMatrix BPM.
* the referenced Business Data project contained multiple Business Object Models
(BOMs)
The error occurred because, in this scenario, the BPM runtime incorrectly
detected incompatible changes, even though the project did not contain any such
changes. For example, if BDP-X contains a case class with an attribute
('attribute1') which is of type 'GlobalY' (which is a global class defined in
BDP-Y), then deploying a new version of BDP-X may fail with an error like this:
...
Incompatible changes made during upgrade of version <number> of model <name>
Details:
The following unsupported changes were made:
1. [Change] Reference eType in attribute1 changed from org.eclipse.emf.
ecore.impl.EClassImpl@227b328a (eProxyURI: #//GlobalY) to org.eclipse.
emf.ecore.impl.EClassImpl@69b15443 (name: GlobalY) (instanceClassName:
null) (abstract: false, interface: false)"
...
In this example, the error suggests that the type of the attribute has changed
from 'GlobalY', even though it has not.
A new onlyCheckNameForReferenceTypesOnUpgrade property is available in the
<CONFIG_HOME>\bpm\<bpm_app_name>\configuration\bds.properties file. You can use
this property to temporarily disable the checking that produces this error and
so successfully upgrade any affected Business Data project:
1. Add the following property to the <CONFIG_HOME>\bpm\<bpm_app_name>\
configuration\bds.properties file:
onlyCheckNameForReferenceTypesOnUpgrade=true
2. Upgrade every Business Data project that is affected by this problem.
3. Change the property value to 'false':
onlyCheckNameForReferenceTypesOnUpgrade=false
WRM-9780 (CN-01483977)
-----------------------------
The Readme did not provide any information about how to determine the version of
the ActiveMatrix BPM application (amx.bpm.app).
This information is now provided. See "How to Determine the ActiveMatrix BPM
Hotfix Version" above.
WRM-9778 (CN-01486185)
----------------------
The ActiveMatrix BPM application now generates additional DEBUG, WARN and ERROR
logging statements relating to Process Engine module deployment. This additional
logging has been added to make it easier to investigate issues that involve
starting, stopping, deploying or undeploying BPM applications.
WRM-9776 (CN-01485111)
----------------------
On a system with the supportTimeZoneInDateTime=false property set (in the
<CONFIG_HOME>\bpm\<bpm_app_name>\configuration\bds.properties file), an attempt
to write a DateTime value to a case object failed, even though the DateTime
value supplied did not include a timezone. An error like this was written to the
BPM log file:
[ERROR] org.hibernate.util.JDBCExceptionReporter - Time-zone not allowed in
DateTime due to supportTimeZoneInDateTime=false setting:
2017-07-07T15:06:25.000Z
NOTE: See WRM-9057 below for more information about the use of the
supportTimeZoneInDateTime setting.
WRM-9763 (CN-01484112, CN-01521719)
-----------------------------------
On a distributed BPM system, an API call using the queryDone operation (in the
ProcessManagerService) returned a Fault response if the paged query was held on
a different node to the one on which the queryDone operation was being executed.
(This occurred because the queryDone operation was incorrectly only trying to
locate the referenced query on the same node.)
WRM-9709
--------
On a DB2 version of ActiveMatrix BPM, the PVM_WORK_ITEM database kept getting
larger over time.
WRM-9692 (CN-01474725)
----------------------
After deploying a new minor version of a case data model, users were unable to
open work items that referenced earlier versions of the case data model. A
BDS_GLOBAL_DATA_READ_CASE_FAULT error was written to the BPM log file:
Fault in reading a BDS case ... Caused by: java.sql.SQLSyntaxErrorException:
ORA-00942: table or view does not exist
This error only occurred in the following scenario:
* Two Business Data Projects were in use - for example, BDP1 and BDP2, and
* BDP1 depended on BDP2, and
* BDP2 contained one or more global classes, but no case classes, and
* multiple versions of BDP1 and BDP2 had been deployed, and
* the BPM node had been restarted since the original deployment of BDP1 and
BDP2.
WRM-9618 (CN-01477393)
----------------------
Deployment of a new minor version of an existing case data project sometimes
failed with an error like this:
Incompatible changes made during upgrade of version <number> of model <name>
This error could occur if you made a change to the case data project that
resulted in the BOM becoming larger - for example, adding a new optional
attribute to a class. The BPM runtime then incorrectly detected incompatible
changes, even though the project did not contain any such changes. (Adding a new
optional attribute to a class is a non-destructive change.)
This error was more likely to occur with very large BOMs, and could be triggered
by any non-destructive change (not just the addition of a new optional
attribute).
WRM-9599 (CN-01475438)
----------------------
When a global signal was thrown by either a service process or a case action,
the signal was not caught by the target business process instance. A
NullPointerException was written to the BPM log file. For example:
java.lang.NullPointerException
at com.tibco.bx.core.signal.BxGlobalSignalApplicationImpl.
getSignalParameters(BxGlobalSignalApplicationImpl.java:613)
.
.
com.tibco.bx.core.BxSystemImpl - dispatchSignal failed with exception
java.lang.IllegalArgumentException: TIBCO-AMX-CF-020001: Fault "{http://
www.tibco.com/bx/2009/management/process}operationFailedFault" not found
in PortType "{http://www.tibco.com/bx/2009/management/process}
processManagement"
WRM-9585 (CN-01473539)
----------------------
ActiveMatrix BPM performs various regular cleanup operations, as scheduled
background tasks, to maintain system state and performance. When using a
distributed ActiveMatrix BPM system, if a network or database outage occurred
some of these cleanup operations stopped running, resulting in a gradual
degradation of system performance (until the BPM nodes were restarted).
WRM-9420 (CN-01453705, CN-01467542)
-----------------------------------
An attempt to cancel a work item or its associated process instance failed if
the work item was in the SUSPENDEDFROMOPENED(14) or SUSPENDEDFROMOPENOFFERED(15)
state. The following error was written to the BPM log file:
[ERROR] - {UT_DELEGATE_BRM_CANCEL_ERROR} - UT-010122: BRM cancel work item
error org.osoa.sca.ServiceRuntimeException: com.tibco.n2.logging.exceptions.
N2LFClassPropertyMappingRunException: {CLASS_PROPERTY_RUN_EXCEPTION} -
ClassPropertyMapping Runtime Failure. Class: [WorkItemModel], Mapping:
[auditedGoRefData], ComponentAttribute: [caseReference], Failed Mapping
Element: [auditedGoRefData]. Exception caused while doing mapping...
WRM-9400 (CN-01311541)
----------------------
If a negative duration was specified in a Reschedule Timer script calculation,
the duration was treated as a positive value and added to the existing deadline
instead of being subtracted from it.
The error occurred because ActiveMatrix BPM does not support the use of negative
durations in a deadline calculation.
With this fix, a negative duration in a deadline calculation is now treated as a
zero-length duration, not as a positive value.
WRM-9386 (CN-01520603)
----------------------
Work items assigned by RQL were occasionally being sent to the Undelivered group
instead of the intended recipient, even though the RQL query was valid.
WRM-9302 (CN-01437559)
----------------------
Completing a work item resulted in the following exception in the BPM log file:
java.lang.ClassNotFoundException: com.tibco.n2.brm.api.OnNotificationDocument
Note: This error only occurred when using an application where:
* a process made a service call to one of the operations provided by the
Business Resource Management services API (brm.wsdl).
* the application version was greater than 1.8.0.
WRM-9090 (CN-00723245)
----------------------
Using a script of the form:
ScriptUtil.subtract(<DateTime1>, <DateTime2>);
in a timer event caused a process instance to hang in an infinite loop on that
timer event, resulting in high CPU utilization by ActiveMatrix BPM. This error
occurred if the <DateTime1> and <DateTime2> values used different timezones.
WRM-8949 (CN-00729540)
----------------------
After migrating a process to a new version, an attempt to resume a previously
halted process instance failed. This problem occurred when the process was
migrated to the new version by setting a migration point on a failed task of the
halted process instance. After migration, on retrying the process instance, the
failed task completed successfully. However, the process instance remained
stuck.
WRM-8944 (CN-00730451)
----------------------
In the ActiveMatrix BPM web service API, using the previewWorkItemFromList
operation allowed the calling user to see the details of a work item that was
currently allocated to another resource in the same offer set, if that work item
had been originally offered to the entity specified in the request.
WRM-8446 (CN-01248735, CN-01249742)
-----------------------------------
TIBCO ActiveMatrix BusinessWorks returned the following validation error when
processing a JMS message containing audit data published by ActiveMatrix BPM:
A validation occurred while parsing: validation error: no declaration for
element "Event" ({com.tibco.xml.validation}
COMPLEX_E_MISSING_ELEMENT_DECLARATION) at /Event[1]
com.tibco.xml.validation.exception.MissingDeclarationException: no
declaration for element "Event"
Messages published in TIBCO ActiveMatrix BPM Event format conform to the
definition of the Event element in the EventCollectorQueryService schema
(ec-basetypes.xsd). However, ActiveMatrix BPM did not include the namespace of
this schema in the "Event" element when publishing an event as a JMS message,
resulting in the validation error reported by TIBCO ActiveMatrix BusinessWorks.
ActiveMatrix BPM now includes the namespace of the EventCollectorQueryService
schema (http://base.api.ec.n2.tibco.com) in the "Event" element when it
publishes the event as a JMS message.
If you want to publish messages without the namespace, add the following (empty)
property definition to the
<CONFIG_HOME>/bpm/<bpm_app_name>/configuration/ec.properties file:
jmsPublicationRootNamespace=
WBPM-1308
---------
After downgrading an ActiveMatrix BPM system from 4.1.0 HF-5 or earlier to
either an earlier hotfix or to 4.1.0, custom client applications with
dependencies on other applications or libraries that were added in the later
hotfixes no longer worked.
For example, suppose you were using 4.1.0 HF-5 and you created a custom client
application which had a dependency on an external library. You added the
external library to App CDN. If you then downgraded the system to 4.1.0 HF-2,
the custom client application would no longer work, as the referenced library
would no longer be in App CDN.
Note: his issue is now resolved and it will not be seen if user downgrades to 4.1.0 HF06 from a later version of Hotfix say 4.1.0 HF07"
WBPM-1306
---------
Part of the startup operation for the ActiveMatrix BPM application (amx.bpm.app)
is to start Application Development, which in turn has to deploy custom client
applications to its runtime repository. Waiting for Application Development
startup to deploy these applications could result in a delay to amx.bpm.app
startup.
A new 'startDeploymentThread' configuration parameter is provided, which allows
you to separate startup of Application Development from startup of the
ActiveMatrix BPM application.
If:
* startDeploymentThread=true: Application Development uses a separate
processing thread to deploy custom client applications. The ActiveMatrix BPM
application can complete its startup without having to wait for custom client
applications to finish deploying. This is the default option.
* startDeploymentThread=false: Application Development startup uses the same
thread as the ActiveMatrix BPM application. The ActiveMatrix BPM application
cannot complete its startup until all custom client applications have finished
deploying. You should set this option if you need custom client applications
to be available as soon as amx.bpm.app starts up.
When startDeploymentThread=true, a series of messages are written to the BPM log
file, showing the thread starting, deploying applications, and completing. For
example:
[ComponentFrameworkDelegate] [INFO ] stdout - 17:01:54.331 INFO
[ComponentFrameworkDelegate] c.t.r.c.RepositoryComponent : Reading properties
from appDev.properties
[ComponentFrameworkDelegate] [INFO ] stdout - 17:02:14.838 INFO
[ComponentFrameworkDelegate] c.t.c.p.c.AppProvisionerService : App
Provisioner starting thread
.
.
[Thread-2744] [INFO ] stdout - 17:02:14.874 INFO [Thread-2744] c.t.c.p.c.
AppProvisioner : Already installed: apps/app-cdn.app.zip
.
.
[Thread-2744] [INFO ] stdout - 17:02:14.875 INFO [Thread-2744] c.t.c.p.c.
AppProvisioner : STARTUP COMPLETED
If a user tries to access Application Development before the thread has
completed, a message indicating that it is still starting is displayed. The
login page displays once the thread has completed.
In ActiveMatrix Administrator, the amx.bpm.app Status changes to RUNNING before
the thread completes.
To enable the use of the startDeploymentThread configuration parameter:
1. Create a new text file, called appDev.properties, in the <CONFIG_HOME>\bpm\
<bpm_app_name>\configuration folder.
2. Copy the following contents into the file:
# Determines if provisioner is enabled, this is required to provision the OOTB
# applications
enabled=true
# Determines if apps are provisioned on a different thread, this allows AMX BPM
# to startup faster but the AppDev applications will not be available until the
# provisioner thread has completed
startDeploymentThread=true
3. Set the startDeploymentThread value to 'true' or 'false' as you require.
4. Save the file.
WBPM-1281 (CN-01489479, CN-01479358)
------------------------------------
In a localized custom client application, localized phrases could be corrupted,
causing them to be displayed as question mark characters ('?') instead of using
the correct, localized characters. The problem was most likely to occur in
applications using non-Latin based languages and could occur if, for example:
* the BPM node was restarted, or
* if a localized phrase was subsequently edited in Application Development
(using Language > VIEW PHRASES)
The problem occurred because system default encoding was used when saving
localization property files, instead of UTF-8. Now, Application Development
always uses UTF-8 encoding when saving localization property files.
NOTE: If you edit and save changes to an exported localization file, either
locally or in a browser-based editor, you MUST use an editor that uses UTF-8
encoding. If you inadvertently save the file in a non-UTF-8 encoding, you may
still encounter this problem if you subsequently re-import the modified custom
client application to Application Development.
WBPM-1297 (CN-01462676)
-----------------------
After deploying a new minor version of a case data model (containing additional
attributes), existing process instances were unable to display the associated
case data. The addition of new attributes should not have impacted existing
process instances.
WBPM-1249
---------
After upgrading the ActiveMatrix BPM system, Application Management sometimes
took a long time to start, or failed to start.
WBPM-1075 (CN-01455768, CN-01256209, CN-01468278)
-------------------------------------------------
When upgrading a version 4.1 ActiveMatrix BPM system to a later 4.1 Hotfix
version, sometimes the existing versions of the example applications hosted in
Application Development were left in place. They should have been upgraded as
part of the upgrade process. An "Already installed" message in the BPM log file
indicated if an application had not been upgraded - for example:
INFO [ComponentFrameworkDelegate] c.t.c.p.c.AppProvisionerService : Already
installed: apps/CaseManagement.app.zip
OS-3505 (CN-00719889)
---------------------
Rendering problems in the Dynamic Organization Configuration dialog made it
difficult to configure a dynamic organization model extension point.
WS-1797 (CN-01520145)
---------------------
When defining a filter, you can now use negative values on integer and decimal
fields.
WS-1794 (CN-01488237)
---------------------
In Workspace, a case view displayed cases from the wrong version. This problem
occurred if multiple versions of the same case class were deployed. These
versions were displayed in the Case View wizard. However, selecting one version
and proceeding to the next page sometimes resulted in filter attributes
belonging to another (unselected) version of the class, so that when the Case
View was created it displayed cases from the wrong version.
WS-1793 (CN-01517761)
---------------------
If custom access was defined for custom menus or toolbars in the userAccess.xml
file, the Configuration Administrator did not display those settings correctly
in the Graphical Editor. (The settings did display correctly in the XML Editor.)
Also:
* Using "Add", "Edit" or "Remove" did not properly update the Graphical Editor
with changes to custom access elements.
* Using "Defaults" did not revert the custom access settings to those in
userAccess.xml.
Issue/Introduction
TIBCO ActiveMatrix(R) BPM 4.1.0 Hotfix 006 is available
Additional Information
TIB_amx-bpm_4.1.0_HF-006_readme.txt