PQ65207: NEW FUNCTION

 A fix may be available

Obtain the fix for this APAR



APAR status
Closed as unreproducible in next release.

Error description
New Function
Local fix Problem summary
****************************************************************
* USERS AFFECTED: All users of WebSphere Application Server    *
*                 V4.0.1 for z/OS and OS/390                   *
****************************************************************
* PROBLEM DESCRIPTION: APAR PQ65207 addresses various problems *
*                      in WebSphere Application Server         *
*                      V4.0.1 for z/OS and OS/390.             *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
The following problems are addressed in APAR PQ65207:

(MD09736) If RRS is cancelled or terminates unexpectedly,
WebSphere for z/OS does not detect the failure and react to it.
This causes WebSphere to exhibit multiple symptoms as it reacts
to the failure in different ways across components.

WebSphere needs to react to an RRS failure quickly and exhibit
the symptoms that improve servicability.

(MD13734) When attempting to deploy an application with a large
EAR file (e.g. larger than 20 Megabytes), the deployment can
fail for a variety of reasons all related to resource or space
shortages. For example, out-of-memory or DB2 table out-of-space
conditions are common.

(MD13763) An exception that is thrown by a bean method may not
be reported in the WebSphere JRAS trace.  The EJB Container's
postinvoke processing produces a trace record with the message:
Bean method threw exception: ...
However, the trace record does print the exception.

(MD13960) When an internally detected problem brings a server
down, the external symptom can be message BBOU0006E with reason
code C2C91236 which indicates that a cancel command was issued
when the user did not issue one. The code issuing the message
does not know where the command came from.  Support needs to be
modified to issue another message, as the internal cancel is
generated, to indicate WebSphere is bringing itself down.
Message BBOU0038W would be appropriate for this, if it was
modified to go to the console.

(MD13992) Abend 0E0 can occur in method
createAndQueueExceptionThrownEvent. In a server defined to write
container SMF records, the second or later method in the
transaction encounters an exception and then program checks
trying to record the fact that the exception occurred. The
program check is because localActivity_Ptr is no longer valid.
The following dump information is presented:

System abend code: 00E0
Abend reason code: 00000029
DUMP TITLE=COMPON=WEBSPHERE Z/OS,  COMPID=5655A9801,ISSUER=BBOR
           LEXT,ABEND IN BBOOSMF /CREATEANDQUEUEEXCEPTIONTH

(MD14116) When a customer specifies DEFER=NO on the LDAP
information in the WebSphere for z/OS Customization dialog, the
SQL generated for the BBOTDBMC job (the one that defines the
ldap database) is missing semi colons on the create index
entries. Error messages similar to the following will be
contained in the BBOTDBMC joblog when it is run:

SQL ERROR DURING EXECUTE IMMEDIATE
DSNT408I SQLCODE = -104, ERROR:  ILLEGAL SYMBOL "CREATE". SOME
         SYMBOLS THAT MIGHT BE LEGAL ARE: DEFER, CLOSE,
         DSETPASS, GBPCACHE, COPY, DEFINE
DSNT418I SQLSTATE   = 42601 SQLSTATE RETURN CODE
DSNT415I SQLERRP    = DSNHSM5X SQL PROCEDURE DETECTING ERROR
DSNT416I SQLERRD    = 0  0  0  -1  217  0 SQL DIAGNOSTIC
         INFORMATION
DSNT416I SQLERRD    = X'00000000'  X'00000000'  X'00000000'
         X'FFFFFFFF'  X'00000000' SQL DIAGNOSTIC INFORMATION

The issue is when the user specifies DEFER=NO on the
WebSphere for z/OS Customization panel, the generated job
BBOTDBMC had missing semi colons on the create index entries.
The generated job BBOTDBMC should be changed in such a way that
it will always have semi colons on the create index entries.

(MD14224) The error message that is displayed during the ear
file processing stage of the BBOWCUPD migration tool is a
generic message that does not accurately reflect the nature
of the problem. The error message displayed needs to be more
specific to enable problem analysis. The following is an
example of the message being issued:

message.1  =  BBON3199E Method processearfile failed.
message.2  =  BBON1200E EARFileProcessor error in task Store
              Component Objects.

(MD14238) When a user defines an environment variable with no
value in AAT/WSAD, the behavior should reflect the behavior
of WebSphere AE in that a naming exception should be thrown
(the environment variable should not even be registered
with naming, WebSphere AE essentially throws it out). WebSphere
for z/OS attempts to get the value of the environment variable
which results in a null-pointer exception.

(MD14239) The JMS Sample Bean can receive exception:

java.lang.ArrayIndexOutOfBoundsException at
com.ibm.ws390.samples.jms.JMSSampleBean.readFromQueue

while reading from the MQ Queue configured for it. The reason
for this is an incorrect operator in a "for loop". Also, the
samples runRequestReply() method can fail if the sample's Queue
is primed with messages.  Customer will experience problems with
the sample if there are any messages on the Queue before sample
execution.
Problem conclusion Temporary fix Comments
APAR 
PQ65206 provides the following problem resolution:

(MD09736) Support has been modified such that when RRS becomes
unavailable, the daemon will terminate causing all control
regions on the system to abnormally end. A critical eventual
action message (BBOU0811A) will be written to the console
indicating that WebSphere must be restarted.

If RRS is unavailable when the daemon is restarted, the daemon
will wait for RRS instead of terminating.

(MD13734) The code which loads, analyzes, and processes the EAR
file was modified to use storage more efficiently. This will
resolve out-of-memory conditions due to inefficient memory usage
within the deployment code.

The DB2 table out-of-space conditions were the result of storing
the entire EAR file in the BLOB field associated with the
BBOMT80_J2EEAPP table.  The code was changed to store a 3-byte
indicator (HFS) in the BLOB field, rather than the entire EAR,
for applications deployed or imported after this fix is applied.
The EAR itself is now stored in the HFS directory for the
application (along with the component parts of the EAR).  The
file name of the EAR in this directory is the application name
followed by .ear, which might be different than the original
name of the EAR file which was deployed.

For applications deployed prior to applying this fix, the EAR
file will remain in the DB2 table, so no migration is required.
Functions such as export which access the EAR file will find and
process the EAR correctly whether it is in the DB2 table or the
HFS.

Though not required, customers who wish to migrate previously
deployed applications so that the EAR is stored in the HFS
rather than the DB2 table can do so by exporting the server
containing the applications, deleting the server, activating the
conversation, then adding a conversation, importing the exported
server, and activating the new conversation.

For all customers, the functions will continue to work the same
way, but EAR files deployed or imported with this fix applied
will be stored in the HFS (in the application directory along
with the components of the EAR) rather than in DB2.  This will
result in decreased space usage for the DB2 tables, but greater
usage in the HFS.

(MD13763) The EJB Container's postinvoke processing now prints
out an appropriate message that includes the exception thrown
from the bean method.

(MD13960) Support has been modified to route message
BBOU0038W to the console.  The path that brings a server down
deliberately has been modifed to issue this message first (in
those cases where it was not already issued).

(MD13992) Module bbooboat.cpp has been updated to get the
activity object pointer from the orb request object.

(MD14116) The DEFER option will be removed from the WebSphere
for z/OS Customization panels.  As a result, the generated job
BBOTDBMC will always have semi colons for the create index
entries.

(MD14224) The error message was changed to indicate the specific
problem encountered while running the tool.

(MD14238) Support has been modified to check if a variable is
null before entering the logic for extracting the value. If it
is the environmental variable entry is deemed invalid and is
ignored.

(MD14239) The "for loop" in readFromQueue was modified to
correct the incorrect operator. The fact that the
Queue needed clearing was accomplished by adding a method,
clearQueue() to the JMS Sample Bean. The client now calls this
method before he does anything else with the bean, besides
creating it.

APAR PQ65207 is associated with SERVICE LEVEL W401400 of
WebSphere Application Server V4.0.1 for z/OS and OS/390.
APAR information
APAR number PQ65207
Reported component name WASKBASE
Reported component ID 5655A9801
Reported release 401
Status CLOSED UR1
PE NoPE
HIPER NoHIPER
Submitted date 2002-08-13
Closed date 2002-09-21
Last modified date 2002-10-02

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:
UQ90052

Modules/Macros
BBOCCPFR BBODBMDT BBOLSMAK BBOMBIND BBOMKDIR BBOODISP
BBOOETDI BBOOMODF BBOOSLPX BBOOWBGN BBOOWCON BBOOWCRD
BBOOWDEL BBOOWDIS BBOOWEDL BBOOWEJN BBOOWELV BBOOWEND
BBOOWQUE BBOOWSEL BBOOWSEM BBOOWTAF BBOPPK BBOPPKI
BBOPSHOW BBOPTCR BBOPUDMP BBOPUDOP BBOPUDOT BBOPUDSC
BBOPUINT BBOPUIOB BBOPUIOP BBOPUIRP BBOPUISC BBORABND
BBORACRW BBORADMP BBORARM BBORAS BBORBACB BBORBGVT
BBORBLOG BBORBOAM BBORBOAX BBORCARM BBORCFMT BBORCLN
BBORCOD BBORCPCE BBORCTF BBORCTT BBORDATA BBORDATU
BBORDEQ BBORDMSG BBORDTF BBORENQ BBOREXIT BBORFRES
BBORFRR BBORGETS BBORJCTF BBORJCTT BBORJLOG BBORJRAS
BBORJSRV BBORJTR BBORJTRC BBORJWTO BBORLCRM BBORLEXT
BBORLOCK BBORLOG BBORLOGI BBORLSCB BBORLSIP BBORMCDP
BBORMCDX BBORMCTB BBOROBML BBOROBMS BBOROMDL BBOROOPI
BBOROSMD BBOROTTI BBORPTRC BBORRDTA BBORRMC BBORRMGR
BBORSDMP BBORSLOG BBORSMCT BBORTBUF BBORTDF1 BBORTDF2
BBORTDL1 BBORTDL2 BBORTFTE BBORTIME BBORTMP BBORTO
BBORTOI BBORTRCD BBORTSRB BBORTSSX BBORTWRT BBOSCSTB
BBOSCURR BBOSDXAR BBOSDXBR BBOSDXCR BBOSDXDR BBOSDXER
BBOSEBL BBOSEC BBOSECCR BBOSECER BBOSECSR BBOSECTF
BBOSECTT BBOSEILI BBOSEILR BBOSEI3I BBOSETUP BBOSEXSI
BBOSEXSR BBOSEXT BBOSICI BBOSICIR BBOSIXIR BBOSLAD
BBOSLAPP BBOSLCL BBOSLCLR BBOSLCTF BBOSLCTT BBOSLDO
BBOSLDOI BBOSLHM BBOSLHMI BBOSLHOI BBOSLLO BBOSLLOI
BBOSLLOR BBOSLMC BBOSLMO BBOSLMOI BBOSLPK BBOSLPRS
BBOSLRA BBOSLRL BBOSLRLI BBOSLSO BBOSLSOI BBOSLUT
BBOSLXL BBOSLXLR BBOSM BBOSNAPP BBOSNCN BBOSNCNR
BBOSNCO BBOSNCOI BBOSNCRI BBOSNCTF BBOSNCTT BBOSNDO
BBOSNDOI BBOSNFED BBOSNLN BBOSNLNI BBOSNLNR BBOSNMO
BBOSNMOI BBOSNNDC BBOSNNDD BBOSNNDM BBOSNNDO BBOSNNDP
BBOSNNSI BBOSNNSJ BBOSNNSK BBOSNNU BBOSNPK BBOSNSM
BBOSNSS BBOSNSSI BBOSNSSR BBOSNUTL BBOSNXN BBOSNXNR
BBOSQAPP BBOSQCK BBOSQCTF BBOSQCTT BBOSQDB BBOSQDBE
BBOSQDBM BBOSQMA BBOSQNP BBOSQOQ4 BBOSQOS BBOSQQE
BBOSQQE3 BBOSQQE4 BBOSQQ ***This field was
truncated. To obtain the full apar
record, please contact your local support
center.***          

Fix information
Fixed component name WASKBASE
Fixed component ID 5655A9801

Applicable component levels
R401 PSY UQ90052    UP02/09/27 P F209

  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


Product categories: Software > Application Servers > Distributed Application & Web Servers > WebSphere Application Server for z/OS
Operating system(s):
Software version: 401
Software edition:
Reference #: PQ65207
IBM Group: Software Group
Modified date: Oct 2, 2002