WebSphere

Portal Express, Version 6.0
Operating systems: i5/OS, Linux, Windows

 

Troubleshoot Document Manager

 

+

Search Tips   |   Advanced Search

 

This section contains information that can assist you in preventing, identifying, and correcting problems related to Document Manager.

 

Search fails with "failed to load" error

Problem: During heavy usage of Document Manager in a data-sharing DB2 database using RRS, the maximum number of concurrent identify level agents can be exceeded. The resulting error message is similar to the following example: 

ExtendedMessage: BBOO0220E: SCHD0104E: The Scheduler poll daemon WPSTaskScheduler (wps/Scheduler)
failed to load tasks from the database due to an exception: java.sql.SQLException: [IBM/DB2]
[T2zos/2.7.72]T2zosReusableConnection.flowConnect:DB2AttachInterface:initRRSAFAttach:1064:RRS 
Identify failed,Return Code=8, Reason Code=X00F30055,Subsystem ID:DS02,Plan Name:, Pklist:NULLID.*DSRA0010E: SQL State = null, Error Code = -99,999DSRA0010E: SQL State = null, 
Error Code = -99,999

Solution: Increase the values in the DB2 DSNTIPE panel as follows: 

IDFORE=100 
IDBACK=1800

 

Blank Private or Submitted Drafts view may display when using document approval workflow with Microsoft SQL Server

Problem: When running under heavy workload using Document Manager document approval workflow with Microsoft SQL Server, users may occasionally see a blank Private or Submitted Drafts view while other users are submitting or deleting drafts. If this occurs, WebSphere Application Server logs show an error message such as java.sql.SQLException: [Microsoft][SQLServer 2000 Driver for JDBC][SQLServer]Could not continue scan with NOLOCK due to data movement.

Solution: Click the Private or Submitted Drafts view icon again to refresh the view. If the error still occurs, try this workaround, which involves two steps:

  1. Apply the SQL Server 2000 post-Service Pack 4 hotfix rollup. This hotfix can be obtained from the following Web page: http://www.microsoft.com/downloads/details.aspx?FamilyID=9c9ab140-bdee-44df-b7a3-e6849297754a&DisplayLang=en.

  2. Add trace flag 9134 as a SQL Server startup parameter. Microsoft fix article KB 815008 discusses this error (601) and introduces a workaround to avoid sending the error to the application. This information is located at the following Web page: http://support.microsoft.com/?id=815008.

The 9134 trace flag is added to alleviate some of these errors under very specific conditions. It is not meant to eliminate them altogether. If the error reoccurs even after applying the suggested workarounds, continue to resubmit the query.

 

Importing large documents in Portal Document may fail with a RollBackException error

Problem: Importing large documents in Document Manager may fail with the following errors in the logs: 

javax.transaction.RollbackException  at com.ibm.ws.Transaction.JTA.TransactionImpl.stage3CommitProcessing(TransactionImpl.java(Compiled Code))
 at com.ibm.ws.Transaction.JTA.TransactionImpl.processCommit(TransactionImpl.java(Compiled Code))
 at com.ibm.ws.Transaction.JTA.TransactionImpl.commit(TransactionImpl.java(Compiled Code))
 at com.ibm.ws.Transaction.JTA.TranManagerImpl.commit(TranManagerImpl.java(Compiled Code))
 at com.ibm.ws.Transaction.JTA.TranManagerSet.commit(TranManagerSet.java(Compiled Code))
 at com.ibm.ws.Transaction.JTA.UserTransactionImpl.commit(UserTransactionImpl.java(Compiled Code))
 at com.ibm.icm.ci.JCRUserTransaction.commit(JCRUserTransaction.java(Compiled Code))
 at com.ibm.icm.ci.JTATransaction.commit(JTATransaction.java:176)
 at com.ibm.icm.jcr.NodeImpl.save(NodeImpl.java:2479)

Solution: Rollback exceptions may be caused by a slow network. Less distance between servers, and more bandwidth between the database and the server, can reduce errors, particularly when dealing with large files in Document Manager.

 

Document Manager plugin is not supported for Apple Mac systems when using the Firefox 1.5 browser

See the topic Supported hardware and software for WebSphere Portal Express Version 6.0, and the section Supported Web browsers, for a list of fully supported browsers.

 

Document Manager table list is incomplete during initial access

Problem: When Document Manager is initially installed, and you view the document library table list of contents, the table might display only icons and no text, or only the document titles with no icons.

Solution: Restart the Document Manager Web application or the WebSphere Portal Express server.

 

Unable to link to Document Manager from Personalization preview or the Personalized List portlet

Problem: Using Personalization preview and the Personalized List portlet, you might be unable to generate links to Document Manager documents under the following conditions:

Solution: Register the document library to a portlet and visit the portlet using an account with the appropriate permissions.

 

Unable to download and install plug-in using Mozilla or Firefox on Linux

Problem: When accessing Document Manager using a Mozilla or Firefox browser on a Linux system, you are prompted to download and install a plug-in. However, when you click the OK button to proceed with the download, nothing happens. This occurs because Java is required to use Document Manager on a Linux system. A Java applet is used to communicate with your local file system. This applet gives you the ability to edit files using your local document editors, as well as to import multiple files and directories from your computer. By default, the Mozilla and Firefox browsers are not configured to use Java.

Solution: Download and install Java on your Linux system before using Document Manager. Java can be downloaded from this Web site: http://www.java.com/en/download/manual.jsp. After installing Java, navigate to the plug-in directory for your browser, For example, /usr/local/mozilla/plugins. Create a symbolic link to the Java plug-in. For example, ln -s /usr/java/jre1.5.0_06/plugin/i386/ns7/libjavaplugin_oji.so.

 

Missing buttons and file tree view

Problem: If you use the XML configuration interface to transfer a portal configuration that includes Document Manager, buttons and the file tree view might be missing from the user interface panels in Document Manager.

Solution: When the Document Manager Web module is redeployed, its groups must be mapped to the All Authenticated group in WebSphere. To map the groups, follow these steps:

  1. Log in to the WebSphere Application Server administrative console.

  2. Select the Document Manager application, as follows:

    1. Expand Applications.

    2. Select Enterprise Applications.

    3. Locate the application that starts with Document Manager.

  3. Click Map Security Roles to Users/Groups.

  4. Assign the Authenticated role to the All Authenticated group.

  5. Save your changes.

  6. Restart the portal.

 

Installation of Document Manager plug-in prevented when using an IPv6 (Internet Protocol version 6) Web address

Problem: Many Document Manager functions depend on the installation of the Document Manager plug-in on your local system. If you use a Firefox browser, and try to access a portal server using an IPv6 URL (Web address), the URL is not accepted correctly, and this prevents installation of the plug-in. An IPv6 URL looks like this: http://[2002:92a:8f7a:20:9:42:81:231]:10038/wps/portal.

A Web site must be accepted as valid or trusted by the browser before installation of the plug-in can be completed. Firefox and Netscape browsers do not accept IPv6 Web sites addresses as valid or trusted. This is due to a bug in these browsers.

Solution: To install the Document Manager plug-in, use an IPv4 URL instead (example: http://yourserver.yourco.com:10038/wps/portal/.

 

Previewing 100MB file might result in an error

Problem: If you attempt to preview a 100MB file in Document Manager, you might see this error message on the preview page: 

The document contents could not be previewed. You must download 
	the document to view the contents.

Solution: Download the document to your local system to preview.

 

Use the timeout feature in DCS for Stellent conversions

Problem: Viewing large-ized files of 100 MB or more in the HTML preview may cause Document Manager to hang.

Solution: The timeout feature can be used to terminate the conversion process so that Document Manager doesn't hang. To enable the timeout feature, go toportal_server_root/shared/app/converters.xml, and in the Converters.xml file, change the TIMEOUT parameter value to 'x' where x is the time in seconds after which the conversion process should terminate.

 

Unable to upload large file (10MB to 500MB) to Document Manager

Problem: Document Manager has the ability to store documents up to 500MB in size. However, there are several factors that limit actually uploading documents this large.

Solution: Adjust the following settings to maximize the ability of Document Manager to upload large documents:

PostSizeLimit setting

You can change the PostSizeLimit setting in the plugin-cfg.xml file, located in the \AppServer\config\cells directory, to allow documents larger than 10MB to be uploaded into Document Manager. If an HTTP server is being used in front of the portal server you will have to modify the PostSizeLimit to upload documents larger than 10MB. The PostSizeLimit value determines the maximum size document that can be uploaded. The default value for PostSizeLimit is 10000000 (10MB). Increase this value to 500000000 (500MB). This is the maximum file size supported by Document Manager.

Transaction timeout value

The transaction timeout value is the maximum duration (in seconds) for transactions to run on the server before an automatic rollback is issued. This could also be an issue when uploading larger files under load. The default value for this setting is 2 minutes. Increasing the value of this timeout will allow larger files to be uploaded. Experiment with different values for this setting to get it right for your configuration. If the transaction timeout value is a problem you will see a message similar to the following in the SystemOut.log: 

WTRN0066W: Transaction (2633625) 
	0001bba9:00000001e0282f991485e7d9a33267078cf208da52b2bc4bd3055b3adf[] 
	has timed out after 120 seconds.
You can modify the value for the transaction timeout value by going to the adminisrtative console and following these instructions:

  1. Select Servers > Application Servers > WebSphere_Portal > Container Services > Transaction Service.

  2. Set the Total Transaction lifetime timeout.

Maximize JVM heap size

Database drivers can read large portions of binaries into memory depending on setup and capabilities. Increasing the maximum JVM heap size on the portal server is recommended to work around this problem. See the Portal tuning guide for recommended upper values. The tuning guide is available from this Web page under the category Performance Guides: http://www.ibm.com/websphere/portal/library.

Other factors that can affect the upload of large documents include:

Browser timeouts

Uploading documents to Document Manager is done via the browser. Depending on the speed of the network connection to the server, and the size of the document, the upload could take a long time. During this time your browser session might time out, resulting in a file upload failure.

Portal server load

The more load the target portal server is under, the more likely it is that timeouts will occur when trying to upload large files. Uploading large documents during off-peak usage hours can help limit this impact.

 

Javascript errors occur when Document Manager plug-in is not installed

Problem: If the Document Manager plug-in is not installed, Document Manager pages are displayed with Javascript errors. Document libraries can be configured to enable or disable installation of the Document Manager plug-in. If you do not want to be prompted to install theDocument Manager plug-in in your browser, you can configure the document library to disable the plug-in installation. However, if you choose not to install the plug-in when prompted, there will be Javascript errors on the Document Manager page.

Solution: To disable the installation of the Document Manager plug-in for a document library, follow these steps:

  1. Log in to Document Manager and navigate to the document library where you want to disable the plug-in installation.

  2. Click on the right-hand drop-down menu and click Configure.

  3. Click to deselect Enable browser plug-in.

  4. Click Save to save the document library configuration.

 

Unable to remove documents from Private Drafts or Locked Documents views

Problem: When a portal user is deleted, then a user with the same name is added to the portal, draft or locked documents created by the deleted user appear in the new user's Private Drafts or Locked Documents view. Because the permissions assigned to the newer user do not apply to the documents created under the deleted user account, the newer user is unable to remove these documents. Details of the document behavior are:

 

Error condition 12030 occurs when importing large files using Domino WebServer 6.5.5

Problem: When importing a file greater than 10 MB in size, using Document Manager and Domino WebServer 6.5.5, the error condition 12030 occurs and the import process fails.

Solution: By default, the Domino Webserver limits the post request size to 10 MB. When importing larger files, the HTTP configuration parameter needs to be modified as follows:

 

Slow Document Manager response time, especially when using workflow

Problem: In a heavily concurrent environment, Document Manager may experience slow response time. This is particularly a problem associated with Document Manager workflow.

Solution: Set the Oracle initialization parameter session_cached_cursors to 100 for the JCR (DB2 Content Manager Runtime Edition) database.

 

Set the JVM max heap size to enable file importing

Problem: When importing files into a document library while running indexing at the same time, or when importing large documents, an out of memory error might appear, or a portal crash might occur.

Solution: During text indexing of imported documents, the binary conversion routine might run out of memory writing out the converted string buffer. This problem is encountered with very large text files, or large files stored in a zip file format such as Word documents or PowerPoint presentations. The file size limit is determined by the size of the JVM heapsize. For example, the typical JVM heapsize is 512MB. A 75MB zip file may expand into over 300MB in size and cause the out of memory error, depending on how much of the heap is being used by other parts of the application. Avoid importing extremely large files into Portal Document Manager.

If import large files, the out of memory error can be avoided by tuning the portal for performance, which involves setting the JVM max heap size limits, as described below:

 

Preview of large documents fails in Document Manager

Problem: When previewing a large document, such as a 10MB or larger document, Document Manager might display out of memory errors, or might stop working.

Solution: Increase the JVM heap size as specified in the troubleshooting item Setting the JVM max heap size to enable file importing. In addition, you should consult the WebSphere Portal Performance Tuning Guide to ensure that the capacity of the entire portal environment is greater than the intended use. Tuning guides for various versions of WebSphere Portal Express are available on the Product Documentation Web site: http://www.ibm.com/websphere/portal/library.

 

Unable to open an Excel document using Internet Explorer 6.0 SP1

Problem: When you download an Excel document stored in Document Manager and select Open, the second time the document is downloaded, the open command might fail. This occurs only in Internet Explorer 6.0 SP1, the version that is included in Windows 2000 and Windows XP prior to SP2. A cache problem in Internet Explorer 6.0 SP1 prevents the call to Document Manager to re-send the Excel document.

Solution: A workaround can resolve the problem temporarily. In Internet Explorer, click on Tools > Internet Options > Delete Files and select Delete all offline content. Restart the browser.

 

Loss of unsupported features upon conversion

Using Document Conversion Services, Document Manager can view and edit many document formats via the document conversion framework. As with any document conversion, however, some loss of formatting and features is expected.

The following list details some reasons why documents viewed using Document Conversion Services do not retain all of the features of the program they were created in:

 

Document Viewer

The viewer makes a read-only HTML rendering of the source document. The business information of the source document should be preserved, though the styles and formatting might diverge from the original. In particular, graphical file formats, such as Freelance Graphics or PowerPoint, render as JPEGs, one per slide. This conversion relies on the graphical capabilities and settings on the server, including:

For best results, users should ensure that the server has the same fonts available as used in the source documents, and that the color depth of the server matches that of the majority of the clients.

 

Rich Text Editor

The Rich Text Editor uses HTML as its document format. The following features are lost in conversion:

 

Problems viewing documents as HTML

Certain file types can be viewed as HTML in Document Manager. If you are having problems viewing a document as HTML, refer to the Viewing documents in HTML topic for a list of supported file types.

 

Problem: HTML view is displayed in black and white

The HTML view of color files in Powerpoint,, Freelance and Word, displays grayscale. In Microsoft Word, images are displayed in black and white, while text is displayed in color.

Solution: The problem lies in starting portal from an XWindows session exported to a Microsoft Terminal Services session. If the portal is started from an XWindows session exported to the Windows box directly, full-color images and text are displayed.

 

Problem: Stellent image conversions not working in Linux for Stellent 8.0.1

Stellent image conversions (HTML view of PowerPoint and Presentation documents, and images in Excel and Word) is not working in Linux for Stellent versions 8.0.1.

Solution: On some Linux installations, particularly newer ones, the Motif libraries that are installed are not compatible with the libraries that are used to build the HTML Export technology. This is known to be the case with most SuSE installations. For example, problems can also be seen when using HTML Export and trying to convert graphics files. For example, zero byte graphics files will be generated if our technology cannot find the proper Motif library. You can check to see if this is the case by running the command ldd libsc_du.so from the portal_server_root/shared/app/oiexport directory.

This command prints a list of the Motif library dependencies If the line for the Motif library looks like the following: libXm.so.1 => not found your system may not have a compatible Motif library installed. The proper solution to this problem is to install a compatible Motif library and use it to build your application. Often, the installation discs for your particular Linux platform will have the proper libraries. If your installation discs do not have the libraries, instructions for downloading a binary rpm can be found at http://www.lesstif.org/download.html. Remember that if you are doing development, you will also need the proper header files, as well.

The following is a list of the Motif library versions used by Stellent when building and testing the outside In binaries: x86 Linux lesstif v. 0.89.0 zSeries Linux lesstif v. 0.92.0 Itanium Linux OpenMotif version 2.1.30. The installation discs for your Linux platform should provide the proper libraries. If the installation discs do not contain the libraries, instructions for downloading a binary RPM be found at http://rpmfind.net/linux/RPM.

 

Problem: Stellent conversion is not working in Linux for Stellent 8.0.1

During Stellent conversion, the following exception is displayed, "java.util.MissingResourceException: Can't find resource for bundle com.stellent.scd.ExportResources, key 127."

Solution: Ensure that the libstdc++-libc6.1-1.so.2 file is in the /usr/X11R6/lib path. Alternatively, you can select the compat-libstdc++-296 while installing the RedHat Linux AS 4.0.

 

Problem: Stellent does not support comments in Word documents

If the user creates a Word document with comments and then uses the document conversion feature to view an HTML version of the document in Document Manager, text content is visible but comments are not.

This is known limitation of Stellent, which is used in the conversion from Word to the HTML view format.

 

Problem: Stellent does not support customized bullet styles

A Word document is created with multiple lines of content, each line using a different custom bullet style, such as telephone or diamond. The document is imported into Document Manager and viewed in HTML format. The custom bullet styles are not preserved in the HTML view.

This is a known limitation of Stellent, which is used in the conversion from Word to the HTML view format.

 

Problem: Stellent does not support automatic update functions in HTML view of a PowerPoint Presentation file

When viewing a PowerPoint presentation file in HTML view, the automatic update functions in the PowerPoint file do not work. This is a known limitation of Stellent and is not supported in Stellent's 8.1 release also.

Limitations viewing documents as HTML
The following issues are expected in the HTML preview of Word and RTF files:

  • Random alignment of text boxes.

  • Preview does not display the following font effects found in Word Shadow, Emboss, Engrave, Double-Strike-Through, Shimmer, or Marching Red Ants.

  • Word text effects such as blinking background, dotted underlines, and sparkling text are not displayed.

  • The symbols such as omega, zeta and so on are not displayed.

  • Multi-column text will be displayed in a single column.

  • Preview does not display table borders, table border colors, styles, page breaks, agates for DBCS characters, or page background colors.

  • Preview does not display paragraph shading or graphical page borders.

  • Vertical text is displayed as horizontal text.

  • Diagrams and tables may not be displayed in the same order as in Word file.

  • Preview displays a paragraph border as individual borders for each line of the paragraph.

  • Text wrapped around a picture in Word is displayed below the picture.

  • Strike-through with small caps may be rendered unevenly.

The following behavior is expected in the HTML preview of a PowerPoint file:

  • Word art text is displayed as plain text.

The following issues can be expected in the HTML preview of an Excel file:

  • Different cell borders not displayed.

  • Preview doesn't work for an empty Excel sheet.

  • Wrapped text in an Excel file looks different in the HTML preview.

The following problems are expected in the HTML preview of a Lotus WordPro file:

  • On all non-Windows platforms, the HTML preview displays only the text content in the WordPro file. Font formats, tables, images, DBCS characters, paragraph styles, and hyperlinks are not displayed.

  • Bitmap images, equations, page breaks, comments, table of authorities, and table of contents are not displayed.

  • Charts are corrupted in the HTML preview.

The following behavior is expected in the HTML preview of a Lotus 1-2-3 file:

  • Diagrams, including ellipses, rectangles, and free lines, are not displayed.

 

Problems with searching specific language documents

Problem: A user cannot search a specific language document in Document Manager. When searching for a specific language document, you may encounter several problems.

Solution:

  1. Create the document with the desired language option. Use the basic search on the same language enabled browser.

  2. You do not have to use the same language version browser. Specify the same language in browser language option or log into a Portal with a user ID that has the same a preferred language.

You can also use the advanced search with the same language option. For example, if you search Japanese documents, you have to create the documents with Japanese language option. Then, you have to search in basic search on a Japanese enabled browser. You can also use the advanced search with the Japanese language option selected.

 

Problems with Document Conversion Services

This section describes the known limitations for Document Conversion Services (DCS). Some of the limitations require you to perform workarounds. The limitations are related to the IBM proprietary formats being converted to different types of files. The conversion of different types of files that are currently supported by DCS are given below:

The Spreadsheet and Presentation Editors no longer exist in the Document Manager and are not supported in the WP 6.0.

Problem: After adding a document type to the content-type.properties file, have to re-import existing documents

Solution: If you find that every time you preview a document in a specific format, you are prompted to download the file, then you need to add or modify the file types in the content-type.properties file. However, for files that already exist in the document library, you will not be able to preview or convert the documents until you import them again.

Problem: Double-byte (DBCS) characters in a document object or diagram display as XXX when a Microsoft Word document is uploaded to Document Manager and viewed as HTML.

Solution: This is a known limitation that will be fixed in the next portal release.

Problem: You might receive the following warning when attempting to use WinZip to open remotedcs.zip:

Warning: the size of the extracted file (28509) does not match the uncompressed size (0) recorded in the zip file.

Solution: Use an archiving utility other than WinZip.

Problem: After adding Open Office file types to the content-types.properties file, and importing Open Office Writer files, cannot preview these files in Document Manager.

Limitation: Document Conversion Services only supports indexing of Open Office documents. Previewing in Document Manager is not supported.

Problem: Correct version of Stellent must be present for Document Conversion Services to work.

Solution: The following table shows the correct version of the viewer for your product:

Product Stellent version
WebSphere Portal, versions 5.0.2_hp, 5.0.3 7.7
WebSphere Portal, versions 5.0.2.3, 5.1.0, 5.1.0.1, 5.1.0.2 7.75
WebSphere Portal version 6.0 8.0.1

To check your Stellent version, look in the file portal_server_root/shared/app/oiexport/common/verbuild.h. The fileversion property shows the version of Stellent being used.

Problem:When a PDF file is imported or converted from a RTE file and viewed in PDM, it doesn't open in the Portal Documentation Manager (PDM).

Solution:Install the latest version of Adobe Acrobat Reader at client side to view the PDF files in the Portal Documentation Manager (PDM).

 

Problems with SearchML

Limitations in SearchML
Page number in the Open Office will not be reflected correctly in SearchML, after conversion.
Double Byte Code Set characters in SearchML are illegible after they are converted from Open Office-Calc format.
Currency format of financial function in Open Office will not be reflected correctly in SearchML after conversion.
Header and footer in Open Office will be lost in SearchML after conversion.
Earlier versions of .doc files created with older applications, such as previous versions of Abiword, may not be converted to SearchML.
PDF files with Chinese characters cannot be converted to SearchML.

 

Known limitations of Rich Text Editor

This section describes known limitations for the default Rich Text Editor. Some limitations require that you perform workarounds.

 

Problem: Pop-up blocker prevents saving documents in the Rich Text Editor

If you have enabled a pop-up blocker in your browser, you will not be able to save an edited document when you click the close button (X) in the Rich Text Editor.

Solution: To avoid this problem, add the portal host site to the list of sites where pop-ups are allowed before you open the document in the rich text editor.

 

Problem: Images copied from the clipboard into the Rich Text Editor are not visible from another computer

When creating or editing a document using the Rich Text Editor, if you copy images from a local application, going through the system clipboard, the images do not appear if you view the document from another computer. The images are only visible from the local machine where the copy and paste occurred. This problem happens because the images are saved with the local URL, and are not saved to the portal server. When the same document is opened on another computer, the images are missing since they are not in the location which is stored in the Rich Text Editor document. This is a known limitation of the Rich Text Editor.

 

Problem: Spell Checker fails to start in a cluster environment after the version upgrade

The SpellCheck functionality is not working after installing or uninstalling WebSphere Portal Express, and the SystemOut.log shows the exception 

448aeb8f ContainerImpl E WSVR0501E: Error creating component null 
	[class com.ibm.ws.runtime.component.ApplicationMgrImpl] 
com.ibm.ws.exception. 
RuntimeWarning: Failed to open  
/opt/WebSphere/AppServer/config/cells/pvcent44Network/applications/
	SpellCheckService.ear/deployments/SpellCheckService at 
com.ibm.ws.runtime.component.ApplicationMgrImpl.start(ApplicationMgrImpl.java:246).

Solution: The error occurs because in WebSphere Application Server, during the installation of an application, there is no way to determine if an application has been fully expanded and is ready to be started. The APAR PQ97212 solves this problem. The users can restart the portal server after waiting for 15 minutes. The exceptions will no longer appear since the application is now been fully expanded and ready to be started.

 

Problem: Recovering a file from the temporary storage space

While working in a Rich Text Editor file, if the portal session is closed prematurely, or the browser is closed accidentally before you can publish your file into Document Manager, it might appear that the file was not saved. However, the file is available in a temporary storage space, which is located in a path similar to this: 

app_server_root\profiles\Profile_Name\temp\WAS_Cell_Name\WebSphere_Portal\odc\
RichTextEditor.war\cds\Session_Id\Filename.ort
You can recover your file from the temporary space. If you need help locating your temporary space, contact the portal administrator.

the document in temporary storage will contain the unpublished content only if it is has previously been saved by clicking the Save button, or clicking the close window X in the upper-right corner and then clicking Save.

 

Use Mozilla, Firefox, or Netscape browsers

 

Problem: Match whole words option not available in Firefox browser

If you edit documents using the Rich Text Editor through the Firefox browser, you will encounter a known limitation when using the Find/Replace function. In the Find/Replace window, the option to Match whole words when searching content is not available.

Solution: Use a different browser.

 

Problem: Symbol fonts are not displayed in Rich Text Editor (RTE) for Mozilla, Firefox, or Netscape browsers

Solution: To display the text in Symbol font, do the following:

  1. In the Rich Text Editor (RTE), change the font to Symbol.

  2. Go to the View Mode.

  3. In the View Mode, change the HTML code for the text with Symbol font. For example, change the following: 

    <span style="font-family: symbol,fantasy;">Text with the symbol font</span>
To <span> <font face="symbol,fantasy;">Text with the symbol font</font></span>

  4. Go back to the Normal View Mode. The text is changed to Symbol font.

 

Problem: Bullets do not display in Rich Text Editor or HTML view when using Firefox browser

By default, the Firefox, browser does not support the Wingdings font which includes bullet characters.

Solution: The workaround is to set the following options for the browser.

  1. Edit the file which sets the encoding of fonts for the browser. The file is called fontEncoding.properties and is located in the [browser]\res\fonts directory. By default the full path is [root drive]:\Program Files\[browser]\res\fonts\fontEncoding.properties.

  2. Add this line to the font encoding file: 

    encoding.symbol.ttf = windows-1252 encoding.wingdings.ttf = windows-1252

  3. Restart the browser.

 

Problem: Must select text first to insert a hypertext link

When inserting a hypertext link into a Rich Text Editor document using FireFox, select some text first, then click on Insert Link to insert the link. The selected text is used as the content for the hypertext link.

 

Problem: Text becomes part of a hyperlink

When text is entered after a hyperlink in Rich Text Editor's HTML Toggle view, using FireFox 1.5 or above, it becomes part of the hyperlink.

 

Problem: Alignment doesn't work for the text with headings that are numbered or bulleted lists

When right or center alignment is applied to the text with headings that are numbered or bulleted lists, using Rich Text Editor in FireFox, only the text moves to the selected aligned position whereas the bulleted or the numbered lists remain left aligned.

 

Rich Text Editor

 

Related to Internet Explorer

 

Swedish special characters are not converted correctly in PDF output

When a file is created using Rich Text Editor and the file contains Swedish characters, then the file is converted to PDF, some Swedish special characters are unreadable.

This is a known limitation of Rich Text Editor. To support other languages when the Rich Text Editor file is output to PDF, font support is required. This font support is currently not available for Rich Text Editor.

 

Problem: Using Format drop down menu changes all words in a line of text

In the Rich Text Editor, using the Format drop down menu changes all words in a line of text regardless of what is selected in the line of text. This is by design.

Solution: If you want to change the text format in part of a line, use the font or size functions instead.

 

Problem: Hyperlinks do not work in edit mode

In the Rich Text Editor, when you are editing a document, the hyperlinks do not work.

Solution: To follow a link, be in view mode.

 

Problem: Cannot cut highlighted text across cells in a table

In the Rich Text Editor, if you highlight text and your cursor crosses the cell border, even if there is nothing in that cell, when you try to cut the selection, the highlighted text will not be deleted.

Solution: Reselect the text making sure you do not cross the cell border.

 

Problem: Characters with extended formats might be lost

In the Rich Text Editor, after you convert a Word document, characters with extended formats such as the ruby feature might be lost.

Solution: All the features of a native application are available so use a native editor instead of Rich Text Editor. Select Native Editor on the Document Manager Edit panel.

 

Problem: Double Byte Character Set font specified in WordPro file not reflected

After converting a WordPro document with Document Conversion Services, the Double Byte Character Set font specified in the WordPro file is not reflected in new file. The default Rich Text Editor font is applied to all the Double Byte Character Set characters in the converted file, while font sizes and font styles, such as bold, are reflected.

Solution: All the features of a native application are available so use a native editor instead of Rich Text Editor.

 

Problem: Cannot save files while in source mode

In the Rich Text Editor, if you save a document while in source mode, when you reopen the document, the source code will display.

Solution: Do not save files while in source mode.

 

Problem: Cannot input Chinese (zh_CN and zh_TW) characters using Input Method Editor in Rich Text Editor

When you create a new Rich Text Editor file and open for editing, you are not able to input Chinese characters using Input Method Editor in Rich Text Editor.

Solution: Click any button or selection in the Rich Text Editor interface before you start the Input Method Editor.

 

Problem: Garbage characters appear in Traditional Chinese

Even if you have set title and description for simplified Chinese, garbage characters appear in Traditional Chinese.

Solution: Perform the following steps to correct this problem.

  1. Login to portal.

  2. Select Administration > Portal Settings > Supported Markups

  3. Select HTML.

  4. Click on Edit Selected Markups

  5. Click on Set locale-specific setting

  6. Select Traditional Chinese.

  7. Click on Edit setting for selected language

  8. Insert UTF-8 as Character Set

  9. Confirm the settings.

 

Problem: Quotation marks are not supported in Rich Text Editor

Quotation marks are not supported in Rich Text Editor on RedHat Linux, pLinux, zLinux and z/OS on all browsers.

Solution: When using Rich Text Editor On RedHat Linux, pLinux, zLinux and z/OS, the following characters are not supported in any of the Rich Text Editor supported browsers:

 

Problem: Mixed mode text direction change is not supported

If the selected text includes both left to right (LTR) text, and right to left (RTL) text, it is called mixed mode text. In Rich Text Editor, mixed mode text direction change (LTR or RTL) is not supported.

Solution: Select text with only one kind of text direction, and then click Text direction left to right or Text direction right to left button.

 

Problem: The 'Tab' function to navigate the cells in a table, is not supported

In Rich Text Editor, if the user presses Tab/Shift+Tab to go to the next cell in a table, it throws a java script error.

Solution: This function is not supported in Rich Text Editor. You can use the mouse to click on the desired cell.

 

Problem: The alt text is not displayed for images

Solution:The users can do one of the following:

Limitations in Rich Text Editor
If you convert a password-protected file into Rich Text Editor file, you will get an error message. Password-protected file conversion is not supported.
When you create a bulleted list, the list is shifted to the right. If you highlight and press the Remove Indent button for one item, it deletes that bullet.
In some locales, if you select text for a numbered list, a list of numbers will appear in front of the text. These numbers will be in Arabic numerals, such as 1, 2, 3, not in the locale's numbering system.
Only .jpg and .bmp images can be displayed in Rich Text Editor HTML preview page.
When a RTE file is converted to Word and viewed in HTML preview, junk characters are appended at the end of the file.
When a RTE file is viewed in HTML preview, some of its content are lost.
The special characters which are stored in two bytes in Single Byte Character Set languages can not be converted from Rich Text Editor file to Lotus WordPro.
After converting a password-protected Lotus WordPro file to Rich Text Editor file you will receive an error message.
The contents of a Rich Text Editor file are not read by the Home Page Reader (HPR) tool.
After converting a Lotus WordPro file to a Rich Text Editor file on Unix, the following can occur:

  • The text features of bold, italic, underline, or a combination of bold, italic, or underline with left, right and center alignment are lost

  • The text features of font colors, font size, font styles and background colors are lost

  • Hyperlinks are changed to plain text

  • Bullets with different styles and formats and numbering with different formats are lost

  • Drop capital letter for the beginning of the paragraph do not display in the correct line of the paragraph

  • Table Of Authorities and Table Of Content do not function

  • Inserted lines in a page without any content in the page are not displayed in the output file. The lines will be displayed only if they are entered between two lines of text. Multiple lines are not reflected between the two texts, only one line is displayed. Different line styles are not reflected as it was in input file.

  • Equations are lost.

  • Background color, object, and chart are lost.

  • Comments inserted for any text are lost.

  • Page breaks are lost.

  • Images are lost.

  • Section divisions are lost

  • If there is a table in the WordPro document, the table format is lost and the contents of the table display vertically.

  • Headers and footers are not appended.

After converting a Lotus WordPro file to a Rich Text Editor file on Windows, the following can occur:

  • Bitmap images, equations, page breaks, comments, table of authorities and table of contents are not converted

  • A WordPro file with content in a non-English language, such as Arabic, Hebrew, or Ukranian, displays junk characters in the HTML view and also after conversion to the Rich Text Editor.

  • Headers and footers are not appended.

  • Charts are corrupted.

After converting a Rich Text Editor file to a Lotus WordPro file, the following can occur:

  • Varying text indentions created using "increase indent" or "decrease indent" feature are changed to a single indention setting in the WordPro file.

  • Space between lines and paragraphs is lost.

  • Right-alignment of paragraphs is lost.

  • Some table cell contents are lost.

  • The inserted comments are not displayed.

  • Default bullet style changes.

  • Images are not reflected.

  • Background colors are not reflected.

  • Hyperlinks are not reflected.

  • Table properties change.

  • Double byte character set displays as garbage characters.

  • Double Byte Character Set Rich Text Editor files cannot be converted.

  • The page breaks inserted in the RTE file are not appended.

  • The text direction changes are not reflected.

  • The multiple lines with numbering in table cells appear in a single line.

  • The charts are not converted.

  • Water mark conversion does not work.

After converting a Microsoft Word file to a Rich Text Editor file, the following can occur:

  • Diagrams displayed in reverse order.

  • Text direction change is not reflected

  • The Symbol font is not supported.

  • Single border shows as multiple borders.

  • Table border color and background color is not reflected.

  • Nested tables do not appear.

  • Random alignment of text boxes

  • Font effects in Word such as, Shadow, Emboss, Engrave, Double-Strike-Through, Shimmer and Marching Red Ants are not converted.

  • Page breaks and agates for DBCS characters are not reflected in the Rich Text Editor.

  • Vertical text is displayed as horizontal text.

  • Text wrapped around a picture in Word is displayed below the picture.

  • Comments inserted for any text are lost.

After converting a Rich Text Editor file to a Microsoft Word file, the following can occur:

  • Hyperlinks are lost.

  • Lines are not displayed.

  • Sixteen background colors are converted and the remaining colors are changed to the default white background.

  • Bullets, numbering, and table borders are not displayed.

  • Page breaks are not appended.

  • Only the .jpeg and .png images are displayed.

  • A table inserted within another table is removed and it's content is displayed vertically in the outer table.

  • Cannot view the Word file when converted from a RTE file that has tables.

  • Comments are lost.

After converting a Rich Text Editor file to a PDF file, the following can occur:

  • Loss of background color.

  • Special characters in Swedish, German, Turkish, and Slovak do not render correctly.

  • Double byte character set displays as garbage characters.

  • If there is .bmp image in a Rich Text Editor file, after conversion to PDF, the HTML view of the PDF is blank. The user cannot view or download the converted PDF file.

  • Applied font styles are not reflected exactly as input; instead, all text content is shown in the same font style.

  • Filled circle bullets are displayed instead of numbered bullets.

  • A document created with all reserved characters fails to convert, giving the error message "An error occurred while trying to convert the document"

  • Change in text direction is not reflected.

  • Empty tables and blank lines are lost in PDF output.

  • Comments are lost in the PDF file.

  • The DBCS characters of other languages such as Chinese or Turkish may not be displayed correctly.

  • The .png image files are lost.

 

Related to Internet Explorer

 

Problem: Choosing the Normal heading in Format inserts HTML tags around lines

In the Rich Text Editor, if you choose Normal for a heading in the Format box, it will automatically insert HTML tags.

Solution: Switch to source mode and manually edit the HTML.

 

Problem: Pressing Enter does not work with the image dialog box

After creating a new Rich Text document, if you use the keyboard sequence of <Ctrl+Shift+p> to bring up the insert image dialog box, tab to the Browse button and then press Enter, nothing happens.

Solution: Invoke the Browse button by pressing the space key, not by pressing Enter.

 

Problem: Cannot select a table to delete

In the Rich Text Editor, you might find it difficult to select a table so that you can delete it.

Solution: To delete a table, highlight the surrounding elements and press Delete or delete the cells one by one.

 

Problem: Up or down arrow keys do not function properly in the Format, Font, or Size drop down lists

In Rich Text Editor, after adding texts or paragraphs, the up or down arrow keys do not function properly in the drop down lists of Format, Font, or Size drop down lists.

Solution: Use the mouse to select the options in the Format, Font, or Size drop down lists.

Limitations in the Rich Text Editor
On Internet Explorer, if you hit Tab, it does not indent. Instead, the dropdown menus in the toolbar are highlighted.
On Internet Explorer, if you insert a table, you cannot undo that action.
On Mozilla, bold, italic, and underline do not work with CTRL+B, CTRL+I and CTRL+U.
The Cut, Copy, Paste buttons does not appear in the Rich Text Editor tool bar if the editor is opened using Mozilla browser.

 

Problem: Using keyboard sequences for copy and paste causes application failure

If you use Ctrl + C to copy data and Ctrl + V to paste data multiple times, you might see all the rows of data moved to one cell.

Solution: Use the drop-down menu to copy and paste instead of the keyboard sequences.

 

Internet Mail Box

 

Problem: Viewing as HTML

In Internet Mail Box, there might be some issues with viewing HTML. For instance,

Limitations
If you e-mail a presentation or a spreadsheet with color charts or graphs using Internet Mail Box, and then view it as HTML, the colors might change.
In the Korean and other locales, titles that have UTF-8 characters in Notes e-mail, might display garbage after being sent via Internet Mail Box.

 

Known limitations of HTML preview

This section describes the known limitations of HTML preview. Some of the limitations require that you perform workarounds.

 

Stellent 8.0.1 compatible platforms

The compatible platforms of Stellent 8.0.1 are given in the following table:

Platform Version
Windows (32-bit) Windows NT 4.0 (and above), Windows 95, Windows 98, Windows 98SE, Windows ME, Windows 2000, Windows 2003, and Windows XP
Windows (64-bit) Windows .Net Server 2003 Enterprise Edition
i5/OS (OS/400 using PASE) V5R2 or later
Red Hat Linux (x86) Advanced Server 2.1, 3.0 and 4.0
Red Hat Linux (Itanium) Advanced Server 2.1, 3.0 and 4.0
Red Hat Linux (zSeries, 31-bit) 7.2
SuSE Linux (x86) 8.x - 9.x and Enterprise Server 9.0
SuSE Linux (Itanium) Enterprise Server 8
SuSE Linux (zSeries, 31-bit) 7 and 8

The Stellent 8.0.1 does not support System p such as AIX/Linux on POWER5 and z/OS environment.

 

Known limitations for Document Manager

This section describes known limitations for Document Manager. Some of the limitations have solutions to help you solve or work around the problem.

 

Problem: Date picker does not work when using Netscape 8.1

When creating a new view in Document Manager using the Creation Date or Last Modified query criteria, the date picker does not pop up if you are working in the Netscape 8.1 browser.

This is a known limitation when using this browser.

 

Problem: Do not use Display like Internet Explorer option in Netscape 8.1

Using the Display like Internet Explorer option in the Netscape 8.1 browser disables the Document Manager plug-in functions. Netscape 8.1 does not recognize installed Internet Explorer plug-ins.

This is a known limitation.

 

Problem: An error message is displayed when first opening a Microsoft PowerPoint file

When a Microsoft PowerPoint file is created or opened for the first time in Document Manager, an error document was not found is displayed. However, the file is opened and the contents are correct. The error message is not shown when the file is opened the second or subsequent times.

This is a known limitation of Document Manager.

 

Problem: Search for documents by author returns no results

When executing a search for documents using the author name as the search criteria, or when creating a custom document view using the document author or modifier as view settings, Document Manager returns no search results, even though there are documents in the library that match the criteria.

Solution: Use lower-case characters when typing the author's name in a document search, or when specifying a document author or modifier while creating a custom document view. If you use the Directory Search Find a Person function to specify an author, and there are upper-case letters in the author name provided by the search results, modify the author name to use lower-case letters.

 

Problem: Failure to create a document or folder in Document Manager

In Document Manager, the maximum allowed lengths of string fields such as title and description can vary depending on the language and database. For English values, all of length limits imposed by the user interface correspond to the maximum allowed database lengths. However, for other languages such as Japanese and Chinese that are represented with more than one byte per character when encoded, the limits imposed by the user interface will be larger than the allowable number of characters in the database. If a user enters more Japanese or Chinese characters than allowed by the database, the user will not be able to create a document or folder in Document Manager.

Solution: To avoid this problem, users might have to enter values shorter than the maximum input allowed by the user interface.

 

Problem: While importing a directory in Document Manager , URLs are reversed in a second Firefox browser window

This problem appears if you are using Document Manager and open two Firefox windows while importing a directory. For example, navigate to the Document Libraries portlet and click Import page for a document library, then click Select Directory and Select and Import. In a second Firefox browser window, enter any URL. The URL is reversed (www.yahoo.com becomes moc.oohay.www).

Solution: This is a known limitation.

 

Problem: An SQL exception occurs when navigating or searching within Document Manager or Personalization on i5/OS

An SQL exception occurs when navigating or searching within Document Manager or Personalization on i5/OS. The error message is SQL statement too long or complex. The message ID is SQL0101.

Solution: IBM DB2® for i5/OS® has a limit in the number of subselects that can be contained within an SQL query. The limit is 31 subselects. The repository used for Document Manager and Personalization executes SQL queries that use subselects. The two primary contributors to the total number of nested subselects are:

Predicates: The number of subselects per predicate depends on how many datatypes include the property referenced in the predicate condition. Each predicate will use approximately 3 subselects at minimum, but that depends on the number of types having the property. So if there are 3 datatypes containing the property there would be 3*3 = 9 subselects used. In the Document Manager and Personalization query and search scenarios, there is typically one datatype containing the referenced property. Therefore if there are 31 subselects possible and 3 used per predicate there are 31/3=10 predicates possible within a query or search. However, this is assuming 0 depth in the hierarchy.

Depth of Hierarchy: There are 2*depth of hierarchy subselects added to the SQL statement in a query or search. Since there is a limit of 31 subselects, there is a maximum depth of 15 that a query or search can execute over (31/2 = 15). However, this is assuming there are no predicates in the query or search statement.

Assuming an average of four predicates per query or search statement for 4*3=12 subselects, that leaves us with 31-12=19 subselects left for query depth. That means a maximum of 19/2=9 levels deep in the hierarchy. An attempt should be made to minimize the depth of the hierarchy for managing content.

 

Problem: Document Manager temporary filespace could fill up server filesystem

The Document Manager temporary filespace gradually fills up with temporary user files. If errors occur and users are not logged out of the portal server during normal operations, the temporary filespace used by Document Manager to hold working files for these users might not be cleaned up. Normally this temporary space is cleaned up when a user logs out of the portal server. The temporary space is located at portal_server_root/InstalledApps/Document_Manager_xx.ear/pdm.war/temp.

Solution: If there are no users logged into the portal server, it is safe to delete all subfolders found in the temp directory. These subfolders contain only transient data and are not needed after a user logs out of the system. Do not delete the files found in the temp directory (for example, empty.ope), only delete folders found in the temp directory.

 

Problem: Importing more than 5000 documents into a document library produces an error

When importing a large number of documents into a document library, an error may occur when the number of imported documents is greater than 5000.

Solution: Divide the import process into several steps so that each import contains less than 5000 documents.

 

Problem: Executing search may return documents based on the meta-data

Meta-data for documents is also indexed for search and executing a search may return documents that do not appear to match the search criteria. This is because the meta-data matched the search criteria.

Solution: There is no workaround to this problem.

 

Problem: Bulk export of document from Document Manager not supported

Document Manager and Document Libraries do not support exporting multiple documents in one action (bulk export).

Solution: This is a known limitation.

 

Problem: People tag drop-down is cut off by the preview window

Using Document Manager with Internet Explorer, if you click on a movie file, such as an AVI or MPEG file, a document preview is provided using a helper application from your local system. The helper application window pops up to ask if you want to save or open the file. If you click Open, the helper application plays the movie in a preview window. Once this window is displayed, if you click the people tag drop-down to see the Last modified by information, the bottom portion of the people tag drop-down is cut off by the preview window, whether there is a movie playing there or not.

Solution: This is a known limitation.

 

Problem: Preview of large document files (10 MB or larger) using Internet Explorer might take a long time to complete

If you preview a document that is 10 MB in size or larger, the Internet Explorer browser might seem to lock up because the preview operation can take 4 or 5 minutes to complete. This is a known limitation of using Document Manager with Internet Explorer.

 

Problem: To initiate plug-in installation, user must click a white, unlabeled button when using Netscape 8.1

When using Document Manager in the Netscape 8.1 browser, the Install Missing plug-in button is shown as a white, unlabeled button. This occurs on Document Manager pages which require the plug-in, such as the document library view, edit a document, or in the Document Libraries portlet.

Solution: Click the white, unlabeled button to display the Install Missing plug-in button in the toolbar. This is a known limitation of using Document Manager with Netscape 8.1.

 

Problem: Importing files with path longer than the maximum path length fails in Windows

Windows has restrictions on the maximum path length allowed for a file. This limit is not known to Document Manager ahead of time. When importing files from Document Manager to your local machine, if the file path is longer than the acceptable Windows maximum path length, then the file import will fail with an error such as: 

Could not download document to file filename. Error code error code.
More explanation of the error codes can be found at the Microsoft Web site: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/debug/base/system_error_codes.asp.

 

Problem: Import or export failure messages reported by the Windows Document Manager plug-in

When using a Windows-based local system to import or export documents from Document Manager, the import or export might fail, and the plug-in will display an error as a result. If the error contains an error code, you can find more information about the exact cause of the error at the following Microsoft Web page: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/debug/base/system_error_codes.asp.

 

Problem: Error message when trying to create a path (subdirectory), or import a file with a length greater than 180 characters

An error message is displayed when the user tries to create a path (subdirectory) with a length greater than 180 characters in Document Manager. The same error occurs if the user tries to import a file into Document Manager when the name of the file is more than 180 characters.

The workaround is to choose a path (subdirectory) name or file name with a length less than 180 characters.

 

Problem: Document does not appear in Submitted Drafts view due to long file path

There are limitations on the overall length of the file path for a document, which includes all folder (directory) names containing the document, plus the file name of the document. The maximum path length allowed in Document Manager is 1000 characters for documents not involved in workflows and 200 for any documents that are involved in workflows. For example, if you create a workflow folder path of 180 characters in length, then a document name of 25 characters, the document is placed in the draft folder instead of the submitted folder, indicating that workflow is broken. If the folder path alone exceeds the 200 character limit, you will not be able to create a document within the workflow. There are several additional folder and file name path length limitations imposed by other components. Due to operating system limitations, creating paths greater than 180 characters in length will cause problems with any operations that interact with the file system, such as roundtrip editing. Also, paths greater than 300 characters (including file name) will cause navigation problems in Internet Explorer as these paths exceed the maximum allowed URL size.

 

Problem: Portal administrator (wpsadmin) can not restrict access of the owner of a PDM file or folder

This is working as designed. There is no way to restrict access for an author of a document.

 

Problem: Some characters cannot be used when naming documents

You can create document names using non-Latin characters (such as Japanese and Cyrillic). However, some characters are invalid, such as asterisks (*), question marks (?), greater than and less than signs (< >), percent signs (%), at signs (@), forward and backward slashes (/ \), apostrophes ('), double quotes ("), pipes (|), colons (:), and square brackets ([ or ]). You receive an error message when trying use a name with these characters. Document names can contain spaces but cannot be the same as an existing document name.

 

Problem: Lotus 123 chart displays twice in HTML view

If you upload a Lotus 123 document to Document Manager in which you have created a chart of any type (for example, line, area, or pie), you might find the HTML view of the document (viewed by clicking on the file name link) displays the chart twice, once within a cell and again below the sheet area.

Comments: This issue will be fixed in a future release.

 

Problem: Colors in charts and graphs change after being e-mailed with Internet Mail Box

If you e-mail a presentation with color charts or graphs, and then view it as HTML, the colors might change.

 

Problem: Receive an error when trying to upload a file with Document Manager

If you try to upload a file or create a document and receive an error message similar to: 

Error creating new document! - resourceNotFound,

your Web server might not be configured to receive HTTP PUT requests or if you use a proxy server, that might not be configured to receive HTTP PUT requests.

Solution: Configure your Web Server to allow HTTP PUT requests. If you use a proxy server, also configure proxy server to allow HTTP PUT requests.

 

Problem: WordPro conversion to Rich Text Editor on Unix servers does not convert graphics

Solution: Configure Portal to use a Remote DCS Server on Windows. On Windows servers, the WordPro import filter is fully-featured, which means that text, character and paragraph attributes, graphics, and other formatting is supported.

 

Problem: On Linux, cannot open and edit files using various editing software and cannot import multiple files and directories from a local system.

Solution:

On Linux, a Java Applet is used to communicate with the client file system. This applet gives you the ability to edit files using your native editors and also to import multiple files and directories from the client file system. In order to run the applet, you will need to have a supported Java Runtime Environment installed on your browser.

Once the Java Runtime Environment is recognized in the browser, you should be prompted to accept an IBM certificate when you first visit the edit or administrative import pages. You should choose to always accept this certificate for the applet to be able to read and copy files from your local machine.

After you accept the certificate, you might need to refresh the page. The plug-in attempts to read and write the files you are editing to the temporary directory defined in the java.io.tmpdir system property. Users need to have authority to create directories and files in that path (usually /tmp/). Lastly, the plug-in requires that KDE be installed and that the kfmclient be in the path. Visit the KDE Web site for information about installing KDE. The kfmclient exec command is used to open files with the editors registered in KDE.

 

Problem: Send a Link is not available when editing a document

When editing a document, the Send a Link option does not appear on the More Actions menu. Save the file and then click on the file name to go to the read view of the document. From the read panel, you will be able to click More Actions and then Send a Link.

 

Problem: Document Manager does not support Web Services for Remote Portlets (WSRP)

An exception occurs when trying to use Document Manager through WSRP.

 

Problem: Searching for terms not in the JVM's locale does not find indexed .html and .txt documents

HTML and text documents are indexed using the JVM's locale and not according to the encoding of the document. Searching for terms not in the JVM's locale does not find indexed .html and .txt documents. For instance, if the system is running an English locale, a DBCS HTML document will be indexed using an English codepage. Searching with a DBCS term will not find the HTML file.

 

Known limitations for Document Libraries

 

Enabling workflow fails due to large number of groups

Problem: If the number of groups is larger than the maximumSearchResults value in the wmm.xml file, then you will not be able to enable workflow. The message "The number of groups are more than supported." is displayed. Also, the Approvers box under Approvals Workflow is empty.

Solution: To work around this error, edit the wmm.xml file and change the default setting maximumSearchResults="200" to maximumSearchResults="300". You can set the maximumSearchResults value to greater than 300 if needed.

 

Authentication errors occur for document import tasks that run longer than two hours

Problem: Authentication errors occur for document import tasks that run longer than two hours When importing large numbers of documents into a document library using the Document Libraries portlet, if the import takes longer than two hours, authentication errors (HTTP 401 errors) can occur.

Solution: The LTPA token that is used during the Document Libraries import task has expired before the import task has completed. By default this value is set to two hours. When doing large document imports that can take longer than the value set for the LTPA token timeout, increase the LTPA token timeout to a value that allows the import job to complete. The LTPA token timeout setting can be found and set in the administration console (Authentication Mechanisms > Timeout under the LTPA section).

 

Importing a file using Document Libraries may fail because of an existing file with the same name

Problem: When importing a file using Document Libraries, if a locked document with the same file name already exists in the target document library, the import fails. While trying to import the file, the portal shows the error "The following document already exists in this library, do you wish to overwrite?" but if you click Yes, rather than successfully overwriting the existing file, the import fails with the message "A document with the same name already exists and you do not have permission to replace it."

Solution: This is a known limitation of this release. You can change the name of the file you wish to import to work around the problem.

 

Create document library fails due to number of groups

Problem: When creating a document library, if LDAP security is enabled, and there are a large number of users (more than 300), the create document library function might fail with the message "The number of groups are more than supported."

Solution: To work around this error, edit the wmm.xml file and change the default setting maximumSearchResults="200" to maximumSearchResults="300".

 

Known limitations for desktop components

 

Problem: Drag and drop of a folder containing documents or subfolders may not copy children

When using drag and drop from desktop components to copy a folder with child documents (or subfolders) from Windows Explorer into a library, the children in the folder (documents or subfolders) might not be copied correctly.

Solution: This is a known limitation of desktop components. The workaround is to copy the documents, or subfolders, one at a time using drag and drop.

 

Problem: Do not use an IPv6 (Internet Protocol Version 6) Web address with desktop components

During desktop component configuration, if you use an IPv6 URL (Web address) in the Document Manager Server URL field, the URL does not work correctly and an error message is displayed. An IPv6 URL looks like this: http://[2002:92a:8f7a:20:9:42:81:231]:10038/wps/desktop.

Solution: Use an IPv4 URL instead (example: http://yourserver.yourco.com:10038/wps/desktop).

 

Problem: Desktop components does not reflect changed document library settings

When document library settings are changed, such as enabling or disabling workflow or document locking, remove the document library and re-add the library so that desktop components can pick up the new settings. No data is lost when removing and re-adding a document library.

 

Problem: Dragging or copying a folder from desktop components to a local drive does not work correctly

Dragging or copying a folder from a Document Manager directory in the desktop components section of Windows Explorer to a local system drive (such as C:) does not copy the folder structure correctly.

Solution: This is a limitation in the design of desktop components. You can copy or drag files directly from Document Manager in your browser to a local directory, so you should use this option instead. Before copying, ensure that a file with the same name does not already exist on the local drive.

 

Prerequisites for viewing images in non-Windows platforms in 6.0 or earlier

To view the images in non-Windows platforms, configure the following:

  1. Connect to the non-Windows machine from your Windows machine using putty or any other Telnet client.

  2. The oiexport directory should be in the PATH as portal_server_root/shared/app/oiexport:$PATH.

  3. Export LD_LIBRARY_PATH /LIBPATH/ SHLIB_PATH to point to theoiexport directory. This environment variable is different for different platforms, as detailed in the table below:

    Platform Library path variable
    AIX LIBPATH
    HP-UX SHLIB_PATH
    Linux LD_LIBRARY_PATH
    Solaris LD_LIBRARY_PATH

  4. Start the XBrowser and enter the host name. Once you login successfully, type echo $DISPLAY at the prompt. This will give you <Windows machineip>:0.0.

  5. In the same Telnet/ssh session where you have exported PATH and LD_LIBRARY_PATH/LIBPATH/SHLIB_PATH, type the command export DISPLAY=<Windows machineip>:0.0. This value will be the same as the value you get when you type echo DISPLAY on an XBrowser terminal.

  6. Change directory to the location where the xclock and xhost files are present, for example, /usr/X/bin.

  7. Type the command ./xclock &.

  8. The Xclock popup comes up on the desktop of the non-Windows machine to which you are connected through X-Browser.

  9. Type ./xhost +.

    This will give you the message access control disabled, clients can connect from any host.

    Changing directory is required because by entering the commands xclock & or xhost +, the error message command not found is displayed.

  10. Change directory to app_server_root/bin.

  11. Restart the server.

  12. Bring up a fresh browser, either Internet Explorer or Mozilla, and try viewing the file containing images.

The above steps are given to start the X server using XBrowser. To start the X server using Exceed and X virtual frame buffer (Xvfb), do the following steps:

  1. Start Exceed.

  2. Connect to the Linux or Linux machine using putty and then type export PATH.

  3. Type export LD_LIBRARY_PATH/LIBPATH/SHLIB_PATH.

  4. Type export DISPLAY.

  5. Type echo $DISPLAY.

  6. Enter xclock & (if xclock appears on the desktop of the non-Windows machine to which you are connected through Exceed or Xvfb, then X server is running).

  7. If xhost + {{ xhost + <jp>} appears, this means conversion works for this particular IP and only + means for all machines.

    When using Linux, launch WebSphere Portal Express from the console or an X Server-enabled client with the same privileges (xhost+) as the console. The graphical conversions require access to an X-Windows server, since they require access to the Xm, Xt and X11 libraries. Also the DISPLAY environment variable must be set to the account that Portal is running under. Note that the DISPLAY must be valid at the point that Portal is started. Otherwise, telneting to a machine and starting Portal from there does not work. You will need to start Portal from an X terminal.

Version 6.0.1 and later

 

Prerequisites for viewing images in non-Windows platforms in 6.01 or later

To view the images in non-Windows platforms, configure the following:

  1. Connect to the non-Windows machine from your Windows machine using putty or any other Telnet client.

  2. The oiexport directory should be in the PATH as portal_server_root/shared/app/oiexport:$PATH.

  3. Export LD_LIBRARY_PATH /LIBPATH/ SHLIB_PATH to point to theoiexport directory. This environment variable is different for different platforms, as detailed in the table below:

    Platform Library path variable
    AIX LIBPATH
    HP-UX SHLIB_PATH
    Linux LD_LIBRARY_PATH
    Solaris LD_LIBRARY_PATH

  4. Start the XBrowser and enter the host name. Once you login successfully, type echo $DISPLAY at the prompt. This will give you <Windows machineip>:0.0.

  5. In the same Telnet/ssh session where you have exported PATH and LD_LIBRARY_PATH/LIBPATH/SHLIB_PATH, type the command export DISPLAY=<Windows machineip>:0.0. This value will be the same as the value you get when you type echo DISPLAY on an XBrowser terminal.

  6. Change directory to the location where the xclock and xhost files are present, for example, /usr/X/bin.

  7. Type the command ./xclock &.

  8. The Xclock popup comes up on the desktop of the non-Windows machine to which you are connected through X-Browser.

  9. Type ./xhost +.

    This will give you the message access control disabled, clients can connect from any host.

    Changing directory is required because by entering the commands xclock & or xhost +, the error message command not found is displayed.

  10. Change directory to app_server_root/bin.

  11. Restart the server.

  12. Bring up a fresh browser, either Internet Explorer or Mozilla, and try viewing the file containing images.
When using Stellent version 8.1.0 and later, the X server dependency has been removed for Linux and Solaris platforms. For these platforms, perform these steps:

  1. Set preferoitrendering to true in the export.cfg file, located in the path portal_server_root/shared/app/export.cfg. An example is provided below: 

    # SCCOPT_RENDERING_PREFER_OIT
   preferoitrendering true

    When this option is set to true, Outside In will use its internal graphics rendering code to produce bitmap output files whenever possible. This option enables Outside In to render images without using the operating system's native graphics subsystem (X11 on Linux and Linux). This option only has an effect if Outside In has been installed with support for both native graphics and internal graphics output solutions.

  2. Make sure that the environment variable GDFONTPATH includes one or more paths to the fonts folder.

    Only TrueType fonts (*.ttf or *.ttc files) are supported. If the variable GDFONTPATH is not found, the current directory is used. If fonts are called for and cannot be found, HTML Export will exit with an error. When copying Windows fonts to a Linux system, the font extension for the files (*.ttf or *.ttc) must be lowercase, or the fonts will not be detected during the search for available fonts. Stellent does not provide fonts with any Outside In product.

 

Parent topic:

Search the product documentation for a solution