Known problems in IBM eDiscovery Manager Version 2.2.1

News


Abstract

This technote contains brief descriptions of problems in IBM eDiscovery Manager Version 2.2.1 that were discovered before the product shipped. Where applicable, workarounds are provided.

Content

1. Exporting file names with non-ASCII characters into a zip file

When you view a zip file that contains file names with non-ASCII characters, use a zip program that supports the viewing of these files such as 7Zip, WinRar, SecureZip, or PKZip. The standard zip viewer that is built into Windows does not support these files, and the files will appear to be missing, although the data is there.


2. Error message: FOLDER_ADD_ITEMS is invalid at this time

The following generic error message is generated in the logs when a scheduled search fails to create a sub-task, or when a long running sub-task from the previous scheduled search is still in the running state. In eDiscovery Manager version 2.2.1, there is no mechanism to distinguish between these two situations.

    [ 2012-10-05 10:38:29.145] [1862360 ] [SEVERE] [780 ] [METHOD] [RecurringWork] [run] Failed to process work item in RecurringWork 2014083084. TaskID {7EF579E8-4F13-4149-8A54-AF318749860A}                          
    com.ibm.icm.edc.ral.casemgmt.CaseServiceInvalidOperationException: Operation: FOLDER_ADD_ITEMS is invalid at this time!

3. Duplicates are removed from advanced case query results

If you are exporting documents from advanced case or folder search (by using the Case or Folder options on the Search tab) and the selected documents contain duplicates, the Total count will include the duplicate documents, but the exported content will not contain duplicates.

The following message is displayed:
    The export completed, but the Total and Successful item counts do not match. This error can occur if the eDiscovery Manager server is stopped while an export task is running. Although the item counts are different, the export results might still be valid.

Verify that the export produced the expected results. If you did not get the results that you expected, try exporting again.


4. Security vulnerability on Windows when DEP is turned off

When eDiscovery Manager is installed in a Windows environment, you can prevent security threats by turning on Data Execution Prevention. Refer to the appropriate Microsoft documentation for your version of Windows:


5. Pausing export tasks can take a long time to complete

Pausing PDF/HTML export tasks can take a very long time to complete. This is because tasks can be paused or stopped only after the current batch completes. For long running tasks, there will be a delay to complete batches. The length of time depends on the stage in which the export task is paused.


6. MSG attachment icons are not displayed correctly

When viewing an email in an exported PST file that has an attachment of type MSG, the icon for that attachment might not be displayed correctly by the viewing application.


7. Server overloads when eDiscovery Manager is restarted and all overdue scheduled searches are started simultaneously

Server overload can occur when eDiscovery Manager is restarted and a large number of scheduled searches are running. If you see this problem, contact IBM Software Support for assistance.


8. Problems with exporting documents

Due to problems in a third-party component of eDiscovery Manager, some items might not be exported properly.
  • IBM Content Collector systems only. Some items of type Files - Content Collector or type Microsoft Exchange email (compound) - Content Collector cannot be exported to PDF format. Attempting to export such items to PDF format produces "Export process failed: access violation" errors in the SystemOut.log file such as the following:
      10/4/10 23:28:09:266 PDT] 00000037               E com.ibm.icm.docviewer.util.DocumentConverter convertFileToFormat Process convertFileToFormat terminated abnormally with exit code 6
      [10/4/10 23:28:09:282 PDT] 00000037               E com.ibm.icm.docviewer.util.DocumentConverter convertFileToFormat Messages from C++: fileID:1101
      Converting to Image.
      Export process failed: access violation
      EXRunExport() failed: access violation (0x03C1)

  • Some Microsoft Word 2002 files that contain macros might fail to export to PDF. When this happens, you can see errors like the following one in the Task Error Details window:

      Error occurred in PDF Extract Plugin while converting the document (whose ID is { documentID }) to PDF format. The document conversion component returned this error: Process convertFileToFormat cannot convert file with OIT file id : n  : Export process failed: access violation.

  • Some Microsoft PowerPoint 2002 documents might fail to export to HTML and PDF. When this happens, you can see errors like the following ones in the Task Error Details window:

      The export plug-in "HTML Extract Plugin" was unsuccessful in converting the item (whose ID is {documentID}) to HTML format. The document conversion component returned this error: Process convertFileToFormat cannot convert file with OIT file id : n  : Export process failed: file is corrupt.

      Error occurred in PDF Extract Plugin while converting the document (whose ID is {documentID}) to PDF format. The document conversion component returned this error: Process convertFileToFormat cannot convert file with OIT file id : n : Export process failed: file is corrupt.

  • Some emails with large attachments might fail to export to HTML. When this happens, an out-of-memory exception such as the following one is displayed:
      10/25/11 10:37 AM    DYRS5054E: The export plug-in "HTML Extract Plugin" was unsuccessful in converting the item (whose ID is A1001001A10C06A10241B40569) to HTML format. The document conversion component returned this error: Process convertFileToFormat canot convert file with OIT file id : 6 : Export process failed: not enough memory for allocation.
      Administrator response: To get more detailed information, search the eDiscovery Manager work manager log file for the keyword "ERROR" near the time that this error occurred. By default, log files are stored in the logs subdirectory of the eDiscovery Manager installation directory, and the default name of the work manager log file is edmworkmgr#.log.

  • In a small number of cases, exported Lotus Notes email that is in Rich Text Format is not properly converted to PDF format.


9. Incorrect error message, installation failure might be due to lack of disk space

In some cases, the following error message is recorded in the eDMInstallDebug00.log file when eDiscovery Manager installation fails due to lack of disk space:
    [wsadmin] Exception caught...
      [wsadmin] **** ERROR **** -  Unable to install the application EDMClient, check the error, it might already exist.
      [wsadmin] WebSphere error:com.ibm.ws.scripting.ScriptingException: com.ibm.websphere.management.application.client.AppDeploymentException: com.ibm.websphere.management.application.client.AppDeploymentException:  [Root exception is java
    .lang.RuntimeException: Deploying /tmp/app9086405967552945284.ear failed.]
      [wsadmin] java.lang.RuntimeException: java.lang.RuntimeException: Deploying /tmp/app9086405967552945284.ear failed.

If you receive this error, verify that you have sufficient temporary disk space to deploy eDiscovery Manager. Check the temporary directory ( %temp% on Windows and /tmp on AIX) and increase the disk space if necessary. For the required disk space, see the system requirements.


10. After restarting a failed export task, the Successful document count might be higher than the Total document count

Discrepancies between the Successful and Total document counts can occur when you are running multiple export tasks and restart an export task that failed with a REPOSITORY_UNAVAILABLE error or a loss of database connection error (for example, db2: ERRORCODE=-4499, SQLSTATE=08001 errors).

The output of the export directories might contain some duplicates, however, no data is lost.

Workaround:
  1. Work with your DBA to fix the database connection issue.
  2. Reduce the number of documents to export and submit the export task again.


1 1. Problems with previewing and exporting plain text files

If the International Components for Unicode (ICU) functionality that is used by eDiscovery Manager cannot determine the character set encoding of a plain text file, then attempts to preview the file in HTML or attempts to export the file to HTML or PDF formats can produce unsatisfactory results.

Workaround:
Instead of previewing or viewing the file in HTML, use the file's native application to view the file. Similarly, instead of exporting the file to HTML or PDF format, export the file to Native or EDRM XML format.


12. Problem with tasks getting stuck in the setup phase

If the work manager loses its connection to the database for any reason during the task setup phase, the task might get stuck and it might not be restartable.

Workaround:
  1. Restart eDiscovery Manager. The stuck task is changed to the "setup failed" state.
  2. Delete the task.
  3. Submit a new task to perform the required work.


13. Exporting Exchange email with large attachments is slow

Exporting Microsoft Exchange email with large attachments by using the Native export format can take a long time. For example, exporting an Exchange email with a 300 MB attachment could take a few hours depending on the operating environment and other factors.


14. Exported collection definitions might contain duplicate CHILDCOMP definitions

When exporting collection definitions from the eDiscovery Manager Administration page, you might see duplicate CHILDCOMP definitions in the exported XML file.

This only occurs in IBM Content Manager collections with child components and multiple itemtypes in the collection. If there are multiple itemtypes but no child collections, or if there is a single itemtype with a child collection or multiple child collections, this problem is not encountered.

Workaround:
Edit the exported XML file by removing the duplicate child components. The revised XML file accurately represents the collection and can be used for archive, backup, or import.


15. Problem with running multiple exports on 32-bit Windows operating system

An out-of-memory error might occur when running multiple exports on 32-bit Windows operating systems. To resolve this issue, try one of the following actions:
  • Upgrade to eDiscovery Manager 64-bit. (recommended)
  • Reduce the number of allowable concurrent export processes and the total number of threads to be used for export.
  • Reduce the Java heap size on 32-bit eDiscovery Manager to 1,300 MB.

Rate this page:

(0 users)Average rating

Add comments

Document information


More support for:

eDiscovery Manager

Software version:

2.2.1

Operating system(s):

AIX, AIX 64bit, Windows 2008 server

Software edition:

All Editions

Reference #:

1614549

Modified date:

2012-11-30

Translate my page

Machine Translation

Content navigation