Release Notes

FileNet Image Services Resource Adapter,

Version 3.2.1 Patch1

June 2007

This document provides last-minute product information, updates to the documentation, and troubleshooting tips. For further troubleshooting and information, see the documentation on the product CD or CSS Web site. This document contains the following sections:

Introduction to 3.2.1 Patch1
New Features in 3.2.1 Patch1
Layout of Product CD
Supported Platforms
Installation and Deployment
Issues Resolved in this Release
Issues to be Addressed in Future
Source Code Sample
Known Software Problems and Restrictions
ISRA Support

Introduction to 3.2.1 Patch1

ISRA is a Resource Adapter for FileNet Image Services (IS). This Resource Adapter is compliant to the J2EE Connector Architecture v 1.0. ISRA is a system-level software driver, which can be used by a Java application component (client) to interact with the IS. ISRA provides an alternative to IDM Web Services for accessing IS in a web application. In addition, it provides a Java based Web solution independent of Microsoft technologies or products.

ISRA version 3.2.1 Patch1 provides additional functionality in comparison to version 3.2. ISRA is packaged in two editions: View and Enterprise.

• View edition supports read-only interactions, namely:
  Logon/Logoff, FindDocuments, GetDocumentContent, GetDocClassIndices, GetMenuValue, IsAnnotated, GetAnnotations, GetPasswordStatus and GetFolderAttributes, retrieving multiple pages of a document in a single call in addition to ChangePassword interaction.
• Enterprise edition supports:
  AddDoc, DeleteDocs, GetDocProperties, UpdateDocProperties, CancelDocPropertiesUpdate, FileDocsInFolder, RemoveDocsFromFolder, GetDocFolders, GetFolderFolders, SaveAnnotations, GetWorkspaces, GetQueues, GetQueuesFields, GetQueueEntries, InsertQueueEntries, DeleteQueueEntries, UpdateQueueEntries, GetSecurityInfo, GetMenuDesc, GetDocClassDesc, PrintDocs, GetPrinterAttributes, CreateQueue, CreateWorkspace interactions , CreateFolders, DeleteFolders, UpdateFolder properties and retrieval of cache list from IS along with all the View edition interactions.

New Features in 3.2.1 Patch1

Following features have been added in ISRA 3.2.1 Patch1 release:

1. Farm or Cluster Support
   Horizontal and vertical clustered environment to support software load balancing.
   Note: ISRA 3.2.1 Patch1 supports clustering only on IBM WAS 5 and 6 Application Servers.
   
   
2. IBM WebSphere Application Server 6 Support
   Support for IBM WAS 6 is provided.
   
   
3. Changes in ISRA Sample Application
   ISRA Sample Application modified to support IBM WAS 6 and load balancing environment (with Session Affinity).
   
   
4. Globalization Support
   This release supports ISRA on non-English environment.
   Note: Following are the restrictions of globalization support:
   • The first character of Workspace must be ASCII character. This is an IS restriction.
   • The locale of the system on which Application Server is running should be the same as of IS locale.
   • The locale of the Web browser and the system running this browser should be the same as of IS locale.
     

Layout of Product CD

Supported Platforms

• BEA WebLogic Server 8.1 SP4 (WLS) on Microsoft Windows 2000/2003 Server, Sun Solaris 9, HPUX 11i, IBM AIX 5.1/5.2 or Linux AS 3.0.
• IBM WebSphere Application Server 5.1/6.0 (WAS) on Microsoft Windows 2000/2003 Server, Sun Solaris 9, HPUX 11i, IBM AIX 5.1/5.2 and Linux AS 3.0.
• JBoss 4.0.3 SP1 Server on Sun Solaris 9, Microsoft Windows 2000/2003 Server and Linux AS 3.0.
• Sun Java System Application Server 8.1.02 on Microsoft Windows 2000/2003 Server, Sun Solaris 9, HPUX 11i, IBM AIX 5.1/5.2 or Linux AS 3.0.
• Oracle 10g on Microsoft Windows 2000/2003 Server, Sun Solaris 9, HPUX 11i, IBM AIX 5.1/5.2 or Linux AS 3.0.
• Sample Application supports Netscape 4.7, Netscape 7.0 and Internet Explorer 6.0.

Installation and Deployment

To deploy ISRA on the target Application Server, follow the steps explained in ISRA deployment guides for the respective Application Servers. To verify the successful deployment, deploy and access the ISRA Sample Application. The Sample Application deployment and configuration is also explained in the ISRA deployment guides. ISRA Programmer's Guide, Installation Guide and Deployment guides for all supported Application Servers are located in the /Docs folder of the product media.

After installation, the installer creates the following files/directories in the installed directory:

Issues Resolved in this Release

The following is the list of issues in ISRA 3.2 Release which have been resolved with Release 3.2.1 Patch1:

• DTS 140955 - INX stubs left in receive mode and not being released on the back-end IS.
• DTS 142545 - Cannot save annotation with different user id.
• DTS 142741 - ISRA fails to read query result from IS with some particular data.
• DTS 145290 - ISRA: Java Viewer - failed to open a document, once added text annotation in Japanese.
• DTS 145483 - The format of "Proprietary" annotations created with ISRA 3.2 contain an extraneous F_VIEWOPTION property.
• DTS 148732 - Problems when using ISRA with container and component managed logon modes at the same time.
• DTS 149561 - ISRA does not show print/fax queues from Application Server.
• DTS 150454 - GetDocumentContent2 does not work as documented.
• DTS 151378 - Daeja Viewer displays "Annotations failed to load" for annot. created by WFD for IS 3.4.2.
• DTS 152061 - F_CUSTOM_BYTES element in "proprietary" class annotation XML is not being saved/retrieved correctly in ISRA 3.2.
• DTS 153017 - GetDocument_content2 causes all the CSM stub configured to be used and be left in RCV mode.
• DTS 154428 - viewerApplet.setAnnotationStampText does not work as desired.
• DTS 154612 - ISRA leaves a queue open after modifying an entry in the queue.
• DTS 157059 - Borders are not visible on annotations created by IDM when displaying an image via the Daeja Viewer.
• DTS 157505 - ISRA 3.2 Pre-ReleasePatch 4: Websphere stops responding after repeated viewings of same document.
• DTS 158539 - ISRA 3.2 - unnecessary overhead calling FN_IS_RPC_Logging.writeLog methods with LoggingLevel set to 0.
• DTS 160491 - Request to make PCH logging configurable in ISRA 3.2.
• DTS 160768 - User cannot print document after "Arguments to COR procedure are incorrect" error occurs.
• DTS 160940 - ISRA Sample application allows saving null and blank space value on required or unique field of Insert Queue Entry page.
• DTS 161054 - Simultaneous logons cause a blank page on ISRA Sample application.
• DTS 161461 - ISRA returns blank instead of NULL if the String column of a queue is NULL.
• DTS 162034 - Security issue when configuring ISRA with LDAP integration.
• DTS 162397 - Receive ErrorCode: FN_IS_RA_10517 and Exception:<77,0,10>The specified object does not exist in the current cache.
• DTS 163241 - ISRA logon using LDAP authentication fails when the user "Display Name" contains a space.
• DTS 163658 - Print overlay text problem (Receive ErrorCode: FN_IS_RA_10517 and Exception :<77,0,10>),when IS is restarted.
• DTS 164072 - Customer's query only returns 16 match set via ISRA sample apps even when there are more.
• DTS 164379 - Support for Non-English environments in ISRA.
• DTS 164740 - Support for Anonymous users using LDAP authentication.
• DTS 164855 - Cannot logon via LDAP if <cn> and <id> tag are different.
• DTS 165917 - ISRA logon fails due to missing sec object id as shown in eladisp.
• DTS 166120 - OutputRecord Validation for some of the Queue Interactions is not done properly in ISRA.
• DTS 166510 - No users can logon to IS through ISRA, while all other IS client products can logon to IS and working fine.
• DTS 167422 - Cannot get correct error code for FindDocuments on ISRA API.
• DTS 167423 - GetQueueEntry returns different error code on ISRA API.
• DTS 167425 - Insert/UpdateQueueEntry returns different error code on ISRA API.
• DTS 167652 - When using GetDocContent2 ISRA is not using the polling interval between each polls.
• DTS 169971 - Annotations are not displayed by Daeja Viewer when environment uses WebSeal.
• DTS 170807 - ISRA: getting error <92,2,8>The user, group, or device object could not be found when trying to retrieve annotations.
• DTS 170866 - Sporadic errors retrieving documents from IS through ISRA.
• DTS 173133 - Explanation on undocumented INX_ANNOT_DOC 52,INX_OTHER_DOC 53,INX_SEP_SHEET 57 in ISRA Programmer Guide.

Issues to be Addressed in Future

The following issues will be addressed in future:

• DTS 133272 - Incorrect format of the Headline Message.
• DTS 123268 - ISRA will not communicate with IS if Print Services are removed or have not been installed. 

Source Code Sample

The FileNet ISRA 3.2.1 Patch1 CD includes a Sample Application. This application was written in Java and is intended to illustrate the programming techniques used with the ISRA.

The Source code for Sample Application is located in the directory where you installed ISRA, e.g., \\fnsw\ISRA321\Sample\Source.

Note: ISRA provides sample code for the purposes of demonstrating how to use the various APIs that are part of ISRA. All of the code that is around the particular API being explained is not suitable for any purpose other than the Sample Application. The assumption here is that the developer who wants to use ISRA has an understanding of Java and how it should be used. As such, we do not make any attempt to make the sample code “scale” or “perform”. No other intent is made or implied, and certainly, no warranty with regard to scalability or performance is being made. This sample should be used strictly as reference material.

Known Software Problems and Restrictions

BEA WebLogic Server 8.1

This section lists the known issues with BEA WebLogic 8.1.
1) A blank page is displayed during login in ISRA 3.2.1 Sample Application.

    Edit the file <WebLogic Installation Directory>\user_projects\mydomain\startWebLogic.cmd (or

   startWebLogic.sh for UNIX platforms)to include ISRA.jar in CLASSPATH environment variable, and restart the

    server. Alternatively, you can set ISRA.jar in the CLASSPATH on the server.

IBM WebSphere Application Server

1. IBM WebSphere 5 and 6 Application Servers destroy the CCI connection stored in HTTP Session object on exit of doPost()/doGet() methods of Servlets or JSPs. To fix this problem in IBM WebSphere 5 apply the IBM WebSphere 5 Fix Pack 2. Please refer http://www.ibm.com/Search?y=11&lang=en&cc=us&q=PQ73966&Search.x=28&Search.y=11 for details on how to apply the fix pack.
   Fix pack for IBM WebSphere 6 is not available.
2. IBM WebSphere Application Server does not provide the crimson implementation of DocumentBuilderFactory class. The Sample Application uses crimson implementation of DocumentBuilderFactory class to implement the Remote printing functionality. Please refer to the IBM WebSphere Deployment guide, which gives detailed steps on how to use the crimson.jar with the Sample Application. For reference, the PMR number for the case opened with IBM is PMR 83866,756,000.
3. IBM WebSphere will not destroy the last connection with FileNet IS. IBM WebSphere creates parked connections because of which one user always remains logged on to IS. To fix this problem in IBM WebSphere 5 apply the IBM WebSphere 5 Fix Pack PK00210. Please refer
   http://www-1.ibm.com/support/docview.wss?rs=180&context=SSEQTP&q1=PQ86338&uid=swg27004936&loc=en_US&cs=utf-8&lang=en for details on how to apply the fix pack.
4. IBM WebSphere Application Server passes the subject as null for container managed sign on, hence ISRA throws an "Invalid Arguments Passed" error with container managed sign on. IBM provided the following links related to this issue: For setting up the context so that the lookup is not cached:
   http://publib.boulder.ibm.com/infocenter/wasinfo/v5r1//index.jsp?topic=/com.ibm.websphere.base.doc/info/aes/ae/rnam_example_behavior.html

   For JNDI caching in WebSphere:
   http://publib.boulder.ibm.com/infocenter/wasinfo/v5r1//index.jsp?topic=/com.ibm.websphere.base.doc/info/aes/ae/cnam_naming_caching.html

5. If multiple ISRA connection factories are configured on IBM WebSphere Application Server 5.0 and each connection factory is configured to use different log files (LogFileName), the multiple connection factories (instances) write to the same log file. This problem occurs if a resource adapter is deployed as a resource. In this case, IBM WebSphere loads multiple connection factories in the same classloader because of which each connection factory uses the same set of classes, thereby creating only one log file. According to IBM, if a resource adapter is deployed as a resource then multiple connection factories (instances) will use the same class loader. The only way to use multiple separate classloaders is to bundle each resource adapter within individual applications (.EAR files). For reference, the PMR number for the case opened with IBM is PMR #82810,004 and DTS created with FileNet is 143515.
6. In case of failover scenarios, there is a possibility that the user may have to perform the operation again to complete the request. For example, in case of AddDoc interaction in the Sample Application, ISRA will throw the "Invalid InputStream object" error if the primary server goes down.

  Note : Please refer to WebSphere DeploymentGuide for configuring multiple connection factories to use separate log files.

ISRA 3.2.1 Patch1

1. If a user is deleted from IS and recreated, the ObjectID of the user will change, however ISRA will continue to hold the invalid ObjectID until the ISRA cache is automatically updated at the CacheRefreshInterval period. In this scenario, interactions involving Security objects might fail until cached meta-data is refreshed. ISRA cache can also be rebuilt by restarting the Application Server.
2. All information (Doc Class information, Users/Groups information, Menu Values, Indices List, Printer Attributes etc.) which is stored in cache
   will be visible only if cache is refreshed (after cache refresh interval) or when Application Server is restarted.
3. A blank page is displayed during login in ISRA 3.2.1 Patch1 Sample Application if the client_helper.jar or ISRA.jar file is not included in Application
   Server's CLASSPATH.
4. ISRA installer uses the JVM bundled within it for the installation. If the installer does not find a platform specific JVM bundled, then it tries
   to launch the setup using the JVM installed on the server. Sometimes user might get an error "No VM found", in that case the workaround
   is to ensure that the bin directory of the installed JDK/JRE on the server is in the PATH. ISRA installer supports JDK/JRE 1.3 and above.
5. WQS services allows user to create a queue in workspace on which user does not have permissions. WQS Service does not return any
   error tuple in this case.
6. WQS Service throws MSSQL exception when user creates a queue with queue name equal to 14 characters in a workspace with name
   equal to 14 characters.
7. Page Foot Note and Time Foot Note are always displayed while faxing documents. A fax operation always displays the page and time
   foot notes and the user cannot set his/her own values for them.
8. If the logged in user changes his/her own password and then logs off from the Sample Application, then user can log into the IS using old
   password until the Application Server destroys the Managed Connection instance. The overlap time can be minimized by setting the unused
   timeout value to a low value or by restarting the Application Server.
9. The retrieval of multiple pages of a document in a single call is implemented in the Sample Application such that the content of the document pages requested for is displayed in the Daeja Viewer and for pages that are not requested, an error page is displayed. The annotation toolbar is enabled on the error page.
10. The implementation of digital signature functionality and COLD/COLD2 documents can be tested using the single page document content retrieval call on the Search/Browse page in the Sample Application. The digital signature and COLD/COLD2 document retrieval does not function as expected with the multi page document content retrieval in a single call. This is a limitation with the Sample Application.
11. Retrieval of multiple pages in a single call (getDocumentContent2) interaction could have performance issues while retrieving content of either large range of pages and/or if the range contains large page size.
12. ISRA requires that the IS servers with the services listed below be available at the start up of an ISRA application:
    • Index Service (INX)
    • Document Service (DOC)
    • Workflow Queues (WQS)
    • Print Services (PRI)
    • Cache Service (CSM)
    • Security Service (SEC)
    • Network Clearing House (NCH)

    If the IS servers are not available, a delay may be experienced during the first ISRA logon. When ISRA is initially started, it will communicate with each of the servers containing these services to verify their existence. If the server does not exist or is on an unavailable network, ISRA will timeout and move on to verify the next server.

13. IBM WebSphere 5 and 6 Application Servers destroy the CCI connection stored in HTTP Session object on exit of doPost()/doGet() methods of Servlets or JSPs. Because of this restriction, a user is allowed to fetch a set of records in a single call. For this, the corresponding changes have been done in the Sample Application. The Next button functionality has been removed, which was used to fetch the next set of records.
14. Bean ID is not being appended with the "InstallLocation" when the properties file for Silent Install is generated. For this, the user needs to simply append the Bean ID "ISRAProductBean" in the ISRASilent.txt file manually.

ISRA Support

If you experience any difficulties with the product media or ISRA, please contact your FileNet Customer Support personnel for assistance.

Note: ISRA is currently supported for production deployment, only in the Managed Environment. You can use ISRA in Non-managed Environment for development purpose only.