Borland Deployment Platform 6.5 Release Notes

Borland Deployment Platform Products

This release note contains information specific to the Borland Deployment Platform 6.5 release. This release of the Borland Deployment Platform 6.5 addresses most of the customer-reported problems and includes the value-added features set.

This release note contains the following sections:

Product Overview

The Borland Deployment Platform includes the following products:

Borland Deployment Op-Center

New features in Op-Center 6.5

Borland Enterprise Server

Platform Extensions 1st June 2005

New features in BES 6.5

BES 6.5 VisiBroker Edition Release Notes

For BES 6.5 VisiBroker Edition Release Notes, go to:

For instructions on migrating a secure VisiBroker application from 5.x and 6.x to 6.5, go to:
Migration to 6.5 VisiSecure for C++

Platform Information

Licensing and Installation instructions

Important: BES and Op-Center must be licensed and registered before they will run; there is no grace period. However, you may start your Management Hub and Agents and then license them remotely using the Management Console. For instructions on how to license your product, go to http://www.borland.com/devsupport/bdpcommon/faq/6.5/licensing/index.html

BES 6.5 installation

For BES installation instructions refer to the Borland Enterprise Server Installation Guide in the following locations:

An installation FAQ is also available at http://www.borland.com/devsupport/bes/faq/6.5/installer/installer.html

Op-Center 6.5 installation

For Op-Center installation instructions refer to the Borland Op-Center (BDOC) Installation Guide in the following locations:

Other Borland and Third-Party Product dependencies

Borland Op-Center and Enterprise Server interoperate with other Borland and third-party developed products such as:

These products can impose restrictions on the range of platforms and OS versions that can be supported by BDOC and/or BES.

Known issues in this release

Borland Enterprise Server known issues

  1. Latest Oracle driver(s), Oracle9i Release 2 (9.2.0.5) (9.2.0.4) (9.2.0.3) and (9.2.0.1), require the Partition be running on JDK 1.4.x. If you require an earlier JDK, then use earlier versions of the Oracle driver.
  2. JDK 1.3.X based applications that do XML transformation must include xalan.jar as part of the archive.
  3. The URL specification of a localhost for Tibco JMS in a Partition's Configuration file must be consistent with the specification of hostname in a DAR file.

    For example, if the localhost name is myhost and you specify localhost:7222 in the jndi-definitions file, you must also specify localhost:7222 for the Tibco Managed Object in the Configuration file. Similarly if you specify myhost:7222 in the jndi-definitions file you must also specify myhost:7222 in the Configuration file. (Case ID: 41633)

  4. There is a known issue in the database level cascade delete support which is only relevant for the "Exclusive" (option-A) commit mode.

    Note: The default commit mode in BES is "Shared" (option B).

    The symptom of this problem is the unexpected occurrence of javax.ejb.ObjectNotFoundException in situations where a transaction involves large numbers of entity beans.

    The work around is to either use the "Shared" commit mode, or to simply increase the value of the entity property ejb.maxBeansInTransactions, or disable transaction cache compaction by setting this property to 0 (zero). (Case ID 41269).

  5. When creating an Apache web server Managed Object, the Managed Object name must not contain any spaces. If spaces are included in the Managed Object name, for example, "Apache Web Server", then the Apache web server will fail to start.
  6. The default security profile shipped is not SSL-enabled.

    If your Partition requires secure transport (EJB container sometimes does), then change the security profile for your Partition to the ssl_enabled profile as follows:

    1. Right-click on Partition and choose Properties.
    2. Go to the Security tab.
    3. Change the profile in the dropdown.
  7. Compiling JSPs using iastool consumes large amounts of memory, and can exhaust a system's virtual memory in many cases. In this case, an out of memory exception will be thrown.

    For example, compiling JSPs in a WAR with 370 JSPs may use as much as 1.5GB of memory. A work around is to increase the size of the virtual memory on your system.

  8. Editing the license.ini file (<install_dir>/var/license.ini) with a Windows-based editor such as Wordpad can add additional trailing spaces that cause your license to become invalid. This issue can be resolved by removing the additional trailing spaces.

    Note: This file should not be edited; there is no valid use case that requires editing of this file.

  9. With network licensing using FlexLM, more than one license may be consumed if there are multiple licensed processes on the same host. Only one license should be consumed in this case.

    The cause of the problem is that a Java-based process (such as the VisiBroker for Java ORB) uses fully qualified host names to check out the license while VisiBroker for C++ processes (such as VisiTransact, or the VisiBroker for C++ ORB) use simple host names to check out the license.

    This problem is expected to be resolved in a future version of FLEXLM server.

    To fix this problem, you must synchronize the full qualified host name and simple hostname as follows:

    1. Log in as root.
    2. Change host name to full qualified host name:

      prompt> hostname "<your_full_qualified_name>"

    3. To check if your change has taken affect, use the following command to print your current hostname:

      prompt> hostname

  10. BES logging is based on the Log4j infrastructure. Some user applications which use Log4j may cause the Partition to hang. User applications should use the per-partition log4j Configuration file, rather than deploying a Configuration file in the archive. By default, this file is located under the Partition's Managed Object footprint at:

    <install_dir>/var/domains/<domain_name>/configurations/<configuration_name>/mos/<partition_name>/adm/properties/logConfiguration.xml

    An additional work around available is to go into the following file:

    <install_dir>/bin/partition.config

    and uncomment the following line:

    vmprop borland.enterprise.server.partition.disableSystemRedirect=true

  11. At the time of this release, there is no integration in JBuilder that supports Borland Enterprise Server 6.5. Therefore, some examples with JBuilder projects and the JBuilder Configurations in the Template Gallery are not supported with any JBuilder release.
  12. HP/UX Itanium systems only: The Apache web server has been known to fail to start if an IP address or hostname is omitted from the Listen directive in the httpd.conf file with the following error message:

    233 Operation not supported: make_sock: for address[::]:7080, apr_socket_set:(IPV6_V6ONLY)

    The work around is to configure the Apache web server Listen directive with a valid IP address/hostname as well as a port number, for example:

    Listen 192.168.10.20:7080

  13. The Apache web server cannot be started manually using the command-line utility if the web server has not first been started using the Management Console (SCU process). This is due to a necessary configuration step that is performed by the Management Console (SCU process) that is not performed by a manual start. The work around for this is to first start the Apache web server using the Management Console. Subsequent manual starts will then be successful.
  14. HP/UX systems only (32 bit and 64-bit): Pet Store application fails to deploy properly when SonicMQ is used as the JMS provider. It works fine with other JMS providers. A patch will be made available later to fix this problem.

Borland Deployment Op-Center known issues

  1. Solaris only: Due to Sun Solaris bug 5045556, the "use user's login environment" Configuration option for a Managed Object does not work when the user's login shell is /bin/sh, /usr/bin/sh, /sbin/sh or /bin/jsh. All other Solaris shells are not impacted. Sun is working on a fix for this problem.
  2. If an Apache web server stops without deleting its httpd.pid file, the Op-Center Agent will not be able to start the Apache web server. This typically happens when the machine crashes while it is running the Apache web server. You must manually delete the httpd.pidfile, which by default is located in: <install_dir>\var\domains\<domain_name>\configurations\<configuration_name>\mos\<apache_managedobject_name>\logs\httpd.pid
  3. When a UNIX text file is downloaded as part of a Managed Object Template, it must follow the UNIX platform rules for Ctrl-M. If the file is edited using a Windows text editor, you must be sure that it does not contain any extra Ctrl-M.

    Watch for this when including .sh files as part of the Custom Executable Managed Object Template.

  4. UNIX only: The Management Agent will be unable to stop a JDataStore server as a Managed Object process unless it is started using the -ui none option. For an example, see the Pet Store cluster.
  5. To allow the Management Agent to manage an IIS server, using the Windows management console, you must set the IISAdmin and WWW services to: Startup Type = Manual. Also be sure all the Recovery actions are set to: Take No Action
  6. If you will be using the Borland Janeva Pet Shop web site with the Pet Store cluster, you have to disable security in the Pet Store security profile.
  7. The Pet Store cluster Template requires an Op-Center license to run.
  8. UNIX only: The first time setuser is run, you must specify a group. For example, setuser -g +o

Known Issues common to BES and Op-Center

  1. If licensing is not configured properly, license files can become corrupt. This is most likely when deploying VisiBroker applications.

    If a licensed component is configured to access a license file on another system, for example, by way of a network mounted drive, the following error will be reported:

    Cannot start: Borland Enterprise Server License error[1023]: Sanctuary Initiation Error :License storage {0} does not apply to current user. Please register your Borland product(s).

    If this occurs, the licence files will be marked as unusable, and you must re-register that license to activate it again.

    As a precaution, after registering your product, you should save a copy of the <install_dir>/var/borland.lic and borland.pkg files to another location.

    Should the license file corruption occur again, licensing can be restored by simply copying these files to the <install_dir>\var folder.

  2. Deleting a license in the License Manager GUI does not fully delete a network license (a license that uses a license server).

    In the License Manager GUI (in the Borland Management Console by way of the Tools|License Manager menu command, you can right-click on a license and delete it. Or using the command line: <install_dir>/bin/lmadm -r info). However, this does not fully delete the license. To delete a network-based license, you must manually delete the regxxx.txt file and/or the concurrent_xxxx.slip file in the <install_dir>/license folder.

    This problem does not apply to deletion of node-locked licenses.

  3. RedHat Linux (Pentium) only: The Installer, when calculating available disk space, displays the following:

    Disk Space Information (for Installation Target):
    Required: 605,149,820 bytes
    Available: Error!

    You should check your available disk space manually at this point. If you have sufficient disk space, installation will proceed normally.

  4. UNIX only: The first time setuser is run, you must specify a group. For example, setuser -g +o
  5. This version of BES includes the latest version of Borland JDataStore viz: JDS 7.02. Please read the Release Notes for JDataStore 7 if upgrading from JDataStore 6 (included in BES 6.0) because the requirements for username/passwords in the database have been tightened.
  6. Right-Click JMX browsing on J2EE appservers (including BES) integrates the open source MC4J Console. There are some known issues of that software that Borland has not attempted to fix and hence passes through as BES/OpCenter problems from the user's viewpoint.
  7. Certain MBeans have operations that are exposed specifically for invocation by other related MBeans. Ideally, these operations should not be invoked from the JMX (MC4J) Console. MC4J Console does not distinguish between these operations and operations that are more relevant to the user coming by way of the Console.

    This is sometimes a problem, such as if the operation requires complex input parameter then you may run through the method invocation wizard and experience a NullPointerException. Even with simple input parameters, you have to explicitly select the argument in the wizard. If not selected, the argument defaults to null and MC4J cannot deal with this.

  8. There is a problem launching MC4J from the Management Console when an external JDK is configured instead of the default JDK. In the case of HP UNIX, an external JDK is always used and this problem is more evident. To fix the problem, edit the mc4j.config file in the <install_dir>/bin directory and change this line to point to the right JDK:

    javahome $var(installRoot)/jdk/jdk1.4.2

  9. MC4J does not work in Management Console only installation. Some jar files required by MC4J are missing from the Management Console only installation. Use the full installation to launch MC4J.
  10. Windows 2003 only: Sometimes when MC4J is launched, a window pops up with the following message:

    "There is no disk in the drive. Please insert a disk into driver E:"

    The work around for this problem is:

    1. After seeing popup window, close it.
    2. In the MC4J console, choose Tools | Options.
    3. Expand system | FileSystems Settings,
    4. Remove both "MyComputer" and "MyHome".
    5. Close all windows and restart the MC4J console.

      The problem should not show up again.

  11. ServerTrace: With ServerTrace activated, the BES Partition may not stop cleanly and appear to hang. This happens due to some non-daemon threads lurking in the ServerTrace subsystem. In such circumstances, we advise you to use the kill operation.
  12. ServerTrace: When ServerTrace is started, it also starts an SNMP agent. This SNMP agent has two port settings. These port settings are specified in the ServerTrace Master configuration file. There could be conflicts between ports used by ServerTrace and other systems on the same host. So, please set the ports to some non-used ports. Default SNMP ports are 161 for the agent and 162 for the traps.
  13. ServerTrace: When the snapshot operation fails, the message displayed does not convey that the operation has failed. Instead, it displays the following message:

    The snapshot named test Snapshot Failed was saved in the snapshots directory

  14. ServerTrace: There is a problem generating a snapshot on a Partition that is configured when another Partition is running.

    The work around is to manually create a directory named "snapshots" in the following directory:

    <install_dir>/var/domains/<domain_name>/configurations/<configuration_name>/mos/partition/<partition_name>

    and then restart the Partition.

  15. ServerTrace: The default snapshot directory "snapshots" gets created when you generate a snapshot for the first time. If the you click on the View Snapshot option without taking a snapshot, the error handling is not clear and the message sent back to you is a CORBA::BAD_PARAM exception.

    Once this exception is encountered, when you generate a snapshot, the snapshot directory does not get created. You have to manually create a directory named "snapshots" in the following directory:

    <install_dir>/var/domains/<domain_name>/configurations/<configuration_name>/mos/partition/<partition_name>

  16. ServerTrace: Enabling and then disabling ServerTrace leaves the ServerTraceMBeans.war deployed in the J2EE server. This WAR contains some initialization code and having this deployed does not have any side effects other than a message in the logs.
  17. ServerTrace: Although everything works fine with ServerTrace configured, sometimes there are entries in the SCU process log like the following: 
    java.lang.UnsatisfiedLinkError: /.automount/smsun8scm1/root/qatest/Optimizeit6.0/OptimizeitEntSuite60/lib/libinstr.so:
        /.automount/smsun8scm1/root/qatest/Optimizeit6.0/OptimizeitEntSuite60/lib/libinstr.so:
        ELF file data encoding not little-endian
  18. When starting a Partition configured with ServerTrace, another benign message in the stderror is: 
    Deserialization of action failed to produce a vector of actions: <install_dir>\var\domains\<domain_name>\configurations\<configuration_name>\mos\<partition_name>\adm\servertrace\default.table
  19. UNIX only: Partition fails to start when ServerTrace is enabled on the Partition. To fix this:
    1. Right-click on the Partition.
    2. Click on Optimizeit | configure.
    3. Select the Master Configuration file and edit this file.
    4. Look for the JDBC tag.
    5. Add ignoreTraining="true"
    6. Save the Master Configuration file and start the Partition. 
      <jdbc enabled="true" reportExceptions="true" reportErrors="true"
reportWarnings="true" reportSoftWarnings="true" ignoreTraining="true"
showOpenResourcesDuringHibernate="true"/>
  20. UNIX only: When starting a Partition configured with ServerTrace, another benign message in the stderror is: 
    "Failed to deserialize the EJB configuration file:<install_dir>/var/domains/<domain_name>/configurations/<configuration_name>/mos/<partition_name>/adm/servertrace/EJBConfig.st
  21. Optimizeit product bug 214623: Attach option does not open with the current view of the server.
  22. Optimizeit product error messages: When configuring ServerTrace/Profiler the Borland Management Console log may have one of the following messages:

    JavaTester error: optit.jar not found in the CLASSPATH

    or

    Can't find OPTIT_HOME, trying cwd

    These are benign messages and do not affect functionality.

  23. Optimizeit product bug 216371: You may receive the following error message when configuring Profiler on more than one BES Partition:

    The port 1470 is already used, the audit system cannot start

    The work around is to use the Audit System Selector utility to explicitly assign a different port number so that the next Partition chooses this new port.

  24. UNIX only: BDOC integration with Weblogic 8.1 SP3 has a Configuration problem. You would see a message "Reason: no weblogicunix1 in java.library.path";. To fix this problem add the following to your Configuration:
    1. For example, for medrec.xml, you would add the following property on the top (where you declare properties): 
      <property name="server.ld.lib.path"
value="${AGENT.weblogic.install.root}/server/lib/solaris"
display-name="Library Path" description="Enter the Library path to use
on the MedRec Server Management Agent" required="true"/>
    2. At the server MO section, under the env-vars element add the following: 
      <env-vars
use-vbroker-env="false" use-current-env="true"
use-default-env="false">
<env-var name="LD_LIBRARY_PATH" value="${server.ld.lib.path}"/>
</env-vars>
  25. Solaris only: ServerTrace 3.0 for WebLogic server is not supported on Solaris.
  26. Linux only: A fully qualified Domain name and IP address need to be the first entry in the hosts file.

    When running a Borland Enterprise Server instance on Linux, the Borland Management Console and iastool are sometimes unable to discover that server when run from a different system. This is because the default Linux hosts file has only the loopback entry 127.0.0.1.

    To correct this, you must add an entry for the local host. For example, on a machine named mylinux that is called mylinux.work.mydomain.com at 172.1.1.1, the following must be added to the hosts file and must be the first entry in that file:

    172.1.1.1 mylinux.work.mydomain.com mylinux

  27. The Borland Management Console and the DDEditor do not function properly in a Management Console-only installation.

    When installing BES or Op-Center, if a full installation is not performed, users may be unable to create new Configurations in the Borland Management Console, and may not be able to run the DDEditor at all. The problem is that some files are missing from the installation.

    The work around for this problem is to reinstall, selecting all product components in the "Choose Product Components" page of the installation wizard, or copy the missing files from a full installation on another machine into the same location in your Management Console-only installation. The files to copy are:

  28. UNIX only: SCU process fails to start with "There may be an Agent running in this footprint..." message.

    This can occur when BDOC/BES has been installed on an NFS filesystem and is being run across NFS. This problem only manifests itself on certain combinations of client and servers on different vendor's UNIX implementations.

    To avoid this problem, it is recommended that you avoid installing BES/BDOC on an NFS filesystem.

Additional copyright information

jUDDI Open Source Project Copyright

Copyright (c) 2003, Steve Viens and contributors. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.