PQ77020: THIS APAR ADDRESSES DEFECTS IN WEBSPHERE APPLICATION SERVER V5.0 FOR Z/OS.

 A fix is available

Obtain the fix for this APAR



APAR status
Closed as program error.

Error description
This APAR addresses defects in WebSphere Application Server
V5.0 for z/OS.
Local fix Problem summary
****************************************************************
* USERS AFFECTED: All users of WebSphere Application Server    *
*                 V5.0 for z/OS                                *
****************************************************************
* PROBLEM DESCRIPTION: APAR PQ77020 addresses various defects  *
*                      in WebSphere Application Server V5.0    *
*                      for z/OS.                               *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
APAR PQ77020 addresses the following defects in
WebSphere Application Server V5.0 for z/OS:

(MD17289) In the customization dialog, when using JES3 to
configure IJP, redundant steps are contained in the jobs
BBOJBSDS and BBOJPAGR.  Running the job BBOJBSDS with the
redundant steps produces messages similar to  the following:

 SY1      IAT6118 0001 CARDS FLUSHED
 SY1      IAT6100 (JOB00010) JOB BBOJBSDS (JOB00042), PRTY=02,
   ID=IBMUSER  NET-ID=*NONE    SUB=JOB00037
 SY1      ICH70001I IBMUSER  LAST ACCESS AT 13:12:20
   ON TUESDAY, JUNE 10, 2003
 SY1      IAT5110 JOB BBOJBSDS (JOB00042) GET
   D FCTMQ CSQ531.WMQX.BSDS0

Similar messages are will be produced when running BBOJPAGR.

(MD17290) Portion of the message MSGS0001E uses ebcdic encoding
and renders unreadable on the console.
Message BBOO0220E MSGS0001E appears as:
Starting the JMS Server failed with exception:
com.ibm.ws390.messaging.QueueManagerException:     @  @     ?

(MD17292) CPU times for Servlets are not collected at
instrumentation points and are presently all populated with
zeros. Support is needed to collect CPU times and pass them to
recently added infastructure.

(MD17297) Using the administrative console, you can get a blank
screen if you try to start cluster member serverx when the node
agent is down. This problem can be seen as follows:

On the administrative console, create a new cluster named
'cluster' to include existing server 'serverx'.  Be sure node
agent is stopped. Navigate to
Servers> Clusters> 'cluster'> Cluster Members.
Select cluster member 'serverx' and click start. A blank screen
will appear. It should stay on the Cluster Members collection
view with a message that the node agent is not started.

The problem was caused by the code in the perform method in
ClusterMemberCollectionAction class.  When the node agent is not
active, the control was forwarded to the
"applicationServerCollection" view which was not defined.

(MD17300) ClassNotFoundException can occur for arbitrary
application classes during server startup. There is a timing
issue when using saveAsFile method such that two files can
be created under the same directory name.

(MD17308) Return code 0x370 from RRS service ATRAFGT.
WebSphere attempted to drive the agent forget service on a
transaction that had already been forgotten.  The URI token
of this forgotten transaction branch was passed to the RRS
service ATRAFGT.  RRS returned code 0x370 which indicates
that the URI token was invalid.  It was invalid because it had
already been forgotten. There was a window when a transaction
timed out between phase one and two of a two phase commit in a
server-as-client environment, when the client would timeout
waiting for the server to commit his branch of the transaction.
When the server did complete the commit, the thread which issued
the commit and the transactional resolver thread both tried to
forget the transaction at nearly the same time.  This caused
the second call to forget (atrafgt) to fail.

(MD17318) An ABEND0C4/ABENDS0C4 in csect BBODASRG in
load module BBODASR when a Control Region ends.

When ENF messages are exchanged on CTL region termination,
a GQSCAN may be done to resync the in storage ASR table to
correctly reflect which servers are up.  If no servers are found
then the code proceeds to process the non-existant results
which leads to an ABEND 0C4.
Also, GQSCAN may not find any active servers due to an erroneous
assumption about the padding of the daemon group name.  This
can result in UNKNOWN OBJECT responses to locate requests to
the daemon when a valid server is active.  This can be detected
by issuing the DISPLAY,SERVERS command to all daemons in the
same cell across a sysplex.  If they do not all show the same
server set active, then this may be the problem.

(MD17324) Using the administration console, you can erroneously
create a cluster with NO servers because the server name field
is not validated. On the administrative console, if you navigate
to: Servers > Clusters; click NEW; enter cluster name;
click NEXT; Leave required server name field blank; click NEXT;
Should get error message saying that Name is required
(as happens when cluster name is left blank). Instead, you get
the next (summary) screen. Click FINISH to create cluster with
NO servers.  A similar problem exists when create new cluster
members.

Implementation assumes that the Apply button is
clicked before clicking the Next button.  There is no checking
whether the Apply button has been clicked before going to the
next panel.  Thus, it caused the problem reported.

(MD17330) Administrative console's start/stop functions for
ListenerPorts are not functional. The impact is that control of
ListenerPorts through the administrative console is lost.

(MD17331) Result of display/modify java trace console commands
go to hardcopy. The need to be routed to the console.

(MD17335) Message "No message text associated w/message
number ADMI2015I" can be received while removing a node that
did not exist. The problem results because
NodeRemovalUtility.java has a typo in the message number. It
should be ADMU2015 instead of ADMI2015.

(MD17341) Cluster member created through wsadmin fails to
initialize. The servant abends with the following:

ABEND IN PC ROUTINE BBOOBIND- ABEND=SDC3
U0000 REASON=04010001  058)

When a cluster member is created through wsadmin, a new entry
is written in the cluster.xml file. However, the cluster
member's uniqueId was not correct. It had the WebSphere
distributed cluster member uniqueId and not the WebSphere
z/OS one. When the transformer ran, it picked the generated
uniqueId from cluster.xml as the server_specific_uuid which
caused the servant to come down.

The root cause of the problem was that the saving of the cluster
configuration data was not taking place. After the
ConfigHelperImpl run to generate the WebSphere z/OS unique code
for the cluster data, it should be saved in the cluster.xml.
Problem conclusion
APAR PQ77020 fixes various defects in WebSphere Application
Server V5.0 for z/OS.

(MD17289) Dialog skeleton BBOJBSDS and BBOJPAGR have been
updated such that the redundant steps will be removed from the
files.

(MD17290) Support in module bbohjms.cpp has been modified to
convert message string to ascii before throwing the exception.

(MD17292) CPU times are collected at instrumentation points and
passed to infactstructure.

Defect fix MD17292 shipped in APAR PQ77020 requires the
following changes to WebSphere V5.0 for z/OS documentation.

The following publication was revised as a result of
defect MD17292:
________________________________________________________________
WebSphere Application Server for z/OS V5
Operations and Administration
SA22-7912-00
________________________________________________________________
NOTE: Periodically, we refresh the documentation on our
Web site, so the changes might have been made before you
read this text. To access the latest on-line
documentation, go to the product library page at:

www.ibm.com/software/webservers/appserv/zos_os390/library.html
________________________________________________________________
Appendix A. Auditing in WebSphere

pg. 136 (new text):
Subtype 7: Servlet activity section

 ffset Offset Name  Length Format  Description
 84   11C  SM120CPU 8  binary  Cpu time in microseconds.

pg. 138-139 (revised text):

Subtype 8: Servlet interval section
 ffset Offset Name  Length  Format  Description
 92   124  SM120WJ4 8  binary  Average cpu time in
                                      microseconds.
 00   12C  SM120WJ5 8  binary  Minmum cpu time in
                                      microseconds.
 08   134  SM120WJ6 8  binary  Maximum cpu time in
                                      microseconds.
________________________________________________________________

(MD17297) Corrected the problem by forwarding to the
"clusterMemberCollection" view when the node agent is
not active.

(MD17300) Support has been modified such that UUID generation
is used to ensure uniqueness of files.

(MD17308) The serialization between the resolver thread and the
syncpoint thread was enhanced to prevent the resolver thread
from forgetting the transaction in this scenario.

(MD17318) Added an 'else' around the GQSCAN result processing
code so as to skip it when there are no results and modified the
assumed pad character for the daemon group name.

(MD17324) Support has been modified to add an "applied" field
to the CreateNewClusterForm and CreateClusterMemberForm.  When
the Next button is clicked, the isApplied() will be checked in
AddServerToClusterAction and CreateClusterMemberAction to make
sure the Apply button has been clicked before going to the next
panel.

(MD17330) In ListenerPortMBeanProxy, modified all the calls to
the ServantMbeanInvoker's method invokeNext to include the name
of the mbean that this proxy is serving. Invocation is thus
routed to the right mbean allowing control and status monitoring
for the selected mbean through the administrative console.

(MD17331) Result of the display/modify java trace console
commands are now routed to the console.

(MD17335) Message number in the NodeRemovalUtility.java has been
changed to ADMU2015. This will result in the text asocciated
with the message being displayed.

(MD17341) The problem is fixed by saving the configuration after
creating the clusterServer and cluster members.

APAR PQ77020 is associated with SERVICE LEVEL W500103 of
WebSphere Application Server V5.0 for z/OS.
Temporary fix Comments
ž**** PE03/08/27 FIX IN ERROR. SEE APAR 
PQ77864  FOR DESCRIPTION
APAR information
APAR number PQ77020
Reported component name WEBSPHERE FOR Z
Reported component ID 5655I3500
Reported release 500
Status CLOSED PER
PE NoPE
HIPER NoHIPER
Special Attention NoSpecatt
Submitted date 2003-07-31
Closed date 2003-08-02
Last modified date 2003-10-08

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Modules/Macros
BBOUBINF          

Publications Referenced
SA22791200        

Fix information
Fixed component name WEBSPHERE FOR Z
Fixed component ID 5655I3500

Applicable component levels
R500 PSY UQ79131    UP03/08/08 P F308

  Fix is available
Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.


Document Information


Current web document: swg1PQ77020.html
Product categories: Software > Application Servers > Distributed Application & Web Servers > WebSphere Application Server for z/OS
Operating system(s):
Software version: 500
Software edition:
Reference #: PQ77020
IBM Group: Software Group
Modified date: Oct 8, 2003