Product: IBM CallPath Enterprise Client
       Version: 6.3.1
         Level: E2106
      Platform: AIX
          Date: 14 September 2001
  Service File: eclaix.tar
       History: eclaix.hst
       Fileset: csebclnt.obj
          Size: 880640   (0.8Mb)
      Checksum: 2641985192
 
 
  Installation Instructions:
  --------------------------
    This Service File is a tar of an installp update fileset.
    It includes both the fileset, csebclnt.obj, and its related .toc file.
    You must have root authority to install this fileset.
     - Download eclaix.tar into a directory.
     - Extract the fileset using the command "tar -xvf eclaix.tar".
     - Start "smit install_update".
     - Select "Install/Update From All Available Software".
     - Enter the directory name to which the fileset was extracted.
     - Enter the fileset name, csebclnt.obj,
         into the field "SOFTWARE to install".
 
    Service files are installed with root ownership and root access permissions.
    If you want users WITHOUT root authority to be able to start or access the
    IBM CallPath Enterprise Client product files,
    the following changes are required:
     - Change the group with the chgrp command for all of the
       directories and files within the installation path.
       For example, to permit the Admin group, issue:
         "chgrp -R Admin /usr/lpp/cseb"
     - Change the file permissions with the chmod command.
       For example, "chmod -R g+x+r /usr/lpp/cseb"
 
 
  Fixes contained in this Service Package:
  ----------------------------------------
 
       Name: APAR IC31537
       Date: 5 Sep 2001
    Library: 9850
    Symptom: On an AIX SMP system a core may occur on the first call to
             TadsConnectToServer. The error report shows a segment fault in
             the C++ library function __DoThrow.
    Problem: C++ try/catch logic was being used in passing return codes. In
             a multithreaded and multiprocessor environment a throw of an
             integer object occasionally produces a segment fault.
   Solution: Remove try/catch logic from the CPE API library (libcseb.a).
             Instead, the conventional means of passing and checking return
             codes is used.
 
       Name: APAR IC31187
       Date: 2 Aug 2001
    Library: 9649
    Symptom: To the user of JTAPI this problem shows up as events not being
             received or as requests timing out. The problem only occurs
             when the JTAPI daemon is running in an AIX system.
    Problem: During the start up of the JTAPI daemon, CSEBJTPI, hundreds of
             TadsSetEventMonitor requests may be issued. While these
             monitors are being issued the JTAPI daemon is not attempting
             to read from its event queue. If enough events are flowing to
             fill up the event queue before all the monitors are completed,
             then the CPE Client may time out the connection associated
             with this event queue and cancel the monitors. In this failure
             situation, the JTAPI daemon does not attempt to re-establish
             the cancelled event monitors.
   Solution: The time out limit used by the CPE Client in sending an event
             is being increased from 5 seconds to 10 seconds. If an event
             send times out, instead of closing the connection, the event
             is discarded but the connection remains open. The following
             text appears in the client log when an event is discarded:
             "Warning: IPCQueue <queue-name> is full ...1 event discarded".
 
       Name: APAR IC30786
       Date: 26 June 2001
    Library: 9500
    Symptom: After a network outage between a Callpath Enterprise client
             and server, an application that reopens a connection and
             registers for events may not receive events for which it is
             registered.
    Problem: The clean up processing associated with a broken Callpath
             Enterprise client to server connection may take over a minute
             to complete (mainly due to slow response from TCP/IP
             requests). If new event monitors are registered during the
             clean up period, they may inadvertently get cancelled by the
             clean up processing.
   Solution: Changed the logic in the Callpath Enterprise client to only
             clean up resources associated with connections that are active
             at the time of the network outage.
 
 
  Files contained in this Service Package:
  ----------------------------------------
    cseb/bin/csebclnt
    cseb/rc/Csebclnt
    cseb/text/csebclnt.copyright
    cseb/text/csebclnt.txt