Rocket® Enterprise Suite

Created On:  22 March 2011 Problem: The Micro Focus supplied version of the IMS REQUATE macro does not produce EQU statements for Access Registers,  a SAVE area DSECT, CSWA_LEN or DWA_LEN.Compile the following Assembler code to show the failure...TEST001  CSECT*         LAM    AR1,AR3,0(AR5)         LA     R1,CSWA_LEN         LA     R2,DLWA_LEN         USING  R8,SAVEAREA         ST     R2,SAVER2*         REQUATE AR=YES,SAVE=YES*         END==============================================================  Micro Focus Assembler version  V3.0.18  07/21/04 Listing as of 03/22/11 21.56   

Created On:  25 March 2011 Problem: A customer received the following error when running a COBOL program: “40 National Language Initialization not set up correctly”. They had set environment variable "LANG=zh_TW.big5" ("big5" represents the Chinese double-byte character set). When they changed the environment variable to "LANG=en_US" the program ran normally. Does Studio Enterprise Edition/Server Express support zh_TW.big5 (or whichever one you require) and how do you find out what is supported? Resolution: The short answer is Server Express/Studio EE does not support LANG=zh_TW.big5. In fact Server Express/Studio EE does not support any Chinese character set.If you have a look in the $COBDIR/lang directory you will see what character sets are supported by Micro Focus. And in this instance there will not be a directory for zh_TW.But, that’s not the cause of this error, and it’s important to understand exactly what the LANG environment variable is looking for.So what do you

Created On:  17 February 2011 Problem: When compiling a CICS program within Server Express, the following error message appeared, even after setting the CICSECM compiler directive:   801-S    External Compiler Module message                 EC9036S MTO ECM licensing error 0006 Resolution: Besides a Server Express COBOL development license, a CICS development license is also required to be installed before you can compile a CICS program.If you purchased the Studio for Mainframe Migration v5.0 (already sundowned) or the Studio for Enterprise Edition v5.1 and later, then there should be two development license keys shipped.  One for the COBOL development license and the other is for the CICS development license.Try to locate the CICS development license and ensure it's installed.  However, if these license keys were not supplied, contact Micro Focus Sales and request a

Created On:  18 March 2011 Problem: Enterprise Server is built in Net Express on Windows . We would like to export the Enterprise Server definition, and then import the definition into our stand alone ES which resides on Red Hat Linux 64-bit.  Is this possible since we are exporting and importing on two different operating systems? Resolution: The work around is to correct the slash, the import imports what is exported and does no conversions. The same happens when you export from UNIX to windows. Incident #2496137 Old KB# 33715#Enterprise#Server#COBOL#StudioEnterpriseEdition#netexpress#EnterpriseServer

Created On:  23 March 2011 Problem: We are receiving the following error when trying to start a region:CASCD0125I Could not exec casmgr32, reason = 0002 Resolution: This is most likely a permissions issue with the /var/mfcobol/es/[region-name] directory.  You will need to delete this directory (and all of it's contents) and start your region again.  This directory is safe to delete as it contains no configuration information, only work files that will be recreated when the region starts. Old KB# 33744#MFDS#EnterpriseDeveloper

Created On:  20 March 2011 Problem: The Host Compatibility Option is used to compile a program containing a Select UNION ALL as shown:EXEC SQL SELECT  DESCR_ERRORE                      INTO   :HV-AFTATER-DESCR-ERRORE                      FROM  AFTATER                     WHERE  BANCA           = :HV-AFTATER-BANCA                       AND  APPLICAZIONE    = :HV-AFTATER-APPLICAZIONE              

Created On:  16 March 2011 Problem: We are trying to launch the MFE database configuration tool in order to activate the HCO feature, but we keep getting an "Unable to find multiple database engines" error message. Resolution: The problem is that the HCO feature has not been installed/activated on the required machine.  Please read attachment HCO.docx on how to check and activate HCO. After following these instructions you will be able to Activate the Database Configuration Tool for MFE 3.1 to select the required database engine - XDB or IBM DB2 for Windows. Incident #2477798 Attachments HCO.docx Old KB# 33695#EnterpriseDeveloper#MFDS#HCODatabaseXDBDB2

Created On:  16 March 2011 Problem: We need to use a Wildcard to check for the existance of Sequential Datasets on the mainframe.  Using the Wildcard symbol '*' to replace part of a Sequential Datatset name generates an error during the synchronisation process. Resolution: The synchronisation process only allows Wildcard allocation for PDS datasets. To use a wildcard for checking for the existance Sequential Datasets on a mainframe the MFDAS command from an Mainframe Express 3.1 command prompt can be used.There is a breakdown of what symbols can be used as wildcards in the MFE 3.1 online help under the Mainframe Access Installation and Usage Guide titled Data Access Services.Examples that can be used with the MFDAS Command are:* means any characters may follow ? means any character in this position means the character in this position must be non-blank - means the character in this position must be blank \\ means the character in this position must be

Created On:  22 March 2011 Problem: A customer is using COMP-5 fields in their mainframe COBOL programs. These are compiled using ENTCOBOL dialect.  They are finding issues due the byte ordering differences (little-endian/big-endian) between MFE on Windows and the mainframe.  I’ve suggested the compiler directive “COMP5-BYTE-ORDER"PORTABLE" as a possible workaround, but the documentation says that this cannot be used with SQL programs.The customer uses XDB, will this be a problem?  If so what would be a solution for the COMP-5 issue? Resolution: This shouldn't be an issue when using the XDB ECM (the Micro Focus SQL precompiler). The ECM runs as another module within the compiler and is made aware by the compiler that the host variables are big-endian when the COMP5-BYTE-ORDER"PORTABLE" directive is in effect.If you are referring to the use of other vendors SQL precompilers, because those run outside of the compiler, they will interpret COMP-5 a

Created On:  16 March 2011 Problem: The first time we compile a program including copybooks on mainframe, compilation is successful.  The second time we compile the same program, the compilation gives an error that the copybook cannot be found. Resolution: This problem can occur if the mainframe datasets/PDS are not suffixed with CPY. To resolve this there is a dataset override in the Server Dataset Configuration.To set this in Mainframe Express 3.1:Select OPTIONS > SERVER ACCESS CONFIGURATION > SORCE CONNECT TAB > CONFIGURE YOUR DATASETSYou can then add your PDS NAMES and CPY as the Extension.  This will now pick up the mainframe copybooks on subsequent compiles with Caching switched on. Incident #2446161 Old KB# 33694#CachingMainframeCompilation#EnterpriseDeveloper#MFDS

Created On:  20 March 2011 Problem: "WSQL007 Error adding line to text editor" error message received when trying to open a large DDL in SQL Wizard. Resolution: The error message received is because the DDL file is too large for the SQL Wizard editor. There is a limitaion on the number of lines that can be opened by the SQL Wizard. It is around 65500 lines. But there is a way to load the DDL and it is to run it as batch through Run Batch menu in SQLWizard. This does not open the text window but does run the SQL statements. Incident #2496850 Old KB# 33724#EnterpriseDeveloper#SQLWizardEditor#MFDS#WSQL007

Created On:  22 March 2011 Problem: When trying to submit a JCL Jobstream to JES, I receive message "JCLCM0164A All job numbers in the range 01000:99999 are in use."  Can this number be increased? Resolution: No, this is not currently possible. If all Job numbers are in use, the only way to allow new Jobs into the system is to delete old output. The maximum JES job number is 99999. Unfortunately, this is a fixed limit and cannot be increased. This gives a real-world limit of over 90,000 jobs.If the system is running out of job numbers,  the recommended approach is to implement a regularly scheduled spool housekeeping job to clean the JES Spool. We suggest that you regularly clear out old JES output from the SPOOL and introduce a method to archive/remove/delete output as soon as it is no longer required on-line. Incident #2469724 Old KB# 33733#EnterpriseDeveloper#MFDS#JESJobnumberJCLCM0164A

Created On:  19 March 2011 Problem: When I tried to run Daemon 1 in Foreground mode I got the following error:"Starting daemon 1 ...Comparison of current locale setting and internal encoding failed. Check error log for details.Enterprise View: ISO8859-1; Environment setting: en_GBDaemon exited because of error. Error starting Daemon 1." Resolution: The message is issued because the Enterprise View 5.7.1 is configured to use National Language Support, i.e. NLS, but the value used by the Linux Environment does not coincide with the one used by Enterprise View product. To disable the National Language Support in Enterprise View 5.7.1, go to Configuration Manager module, open Work With Environment Properties and select the COMMON settings.Change the NLS_ENCODING property to NONE. The default value used is ISO8859-1.If you wish to use the National Language Support with Enterprise View 5.7.1, then the following environment variables have to be exported in your Linux pro

Created On:  19 March 2011 Problem: When doing an impact analysis in Enterprise View for a variable, the point at which the attributes is populated (Input - MOVE) is identified, however that attribute is then referenced in the VALUE parameter of a SQL INSERT. Its reference in that statement is not identified. Is this how the tool is expected to operate? Resolution: The product discovers also the references to the SQL Inserts. If the value of a variable is passed to a SQL Insert, the product will display this reference. The misleading thing is that it shows the starting row of the SQL Insert structure and not the exact row where in the Insert the variable participates. In the two documents attached to this KB article, there is an example of how to achieve the desired result and retrieve the impacted candidates in Enterpirse View 5.7.1 and Enterprise View 5.6.2. In Enterprise View 5.7.1 there are improvements in the impact analysis module that is why there is a

Created On:  23 March 2011 Problem: We are getting the following error when starting an Enterprise Server region:CASCD0003S Failed to 'exec' COBOL RTS, reason = 13 Resolution: Reason 13 is a unix errno 13 which is Permission Denied. This is due to wrong permissions on the CAS executables in the Server Express installation.As root, execute $COBDIR/bin/caspermAnswer the questions and when asked about the Enterprise Server administrator userid, specify the userid that was being used when trying to start the region with. This will set the proper permissions and that userid will be able to start regions. Old KB# 33746#StudioEnterpriseEdition#Server#Enterprise#EnterpriseServer#COBOL#ServerExpress

Created On:  23 March 2011 Problem: We encountered the following problem when we compiled a program that uses NEXT VALUE expression for a SEQUENCE running HCO in Microfocus Mainframe Express, version 3.1 WrapPack 08.COBDB0100S SQL0104N  An unexpected token ":WSN-NUM-PRG      = (NEXT VALUE FOR IPSEQBNF)" was found following "SET ".  Expected tokens may include:  "".  SQLSTATE=42601   : "C:\\mfuser31\\Incidents\\Problema su SEQUENCE\\TESTSQL\\Source\\Testsql2.CBL"(41,19)Compilation complete with 1 errors, 0 warnings, 0 notices and an exit code of 12The precompiled  fails to interpret the SET command on a SEQUENCE.        EXEC SQL        SET :WSN-NUM-PRG = (NEXT VALUE FOR IPSEQBNF)      END-EXEC.We use the following environment:             DB2 Client 9.5  on the local m

Created On:  19 March 2011 Problem: A program used as starting point. In our case this is ‘RPLB071A’. This program is called in PROC – ‘RPLB071A’ which in turn is called in JCLs – RPL1267D, RPL1367D, RPL2167D, RPL2367D, RPL3430D. In our environment, when the program(object) RPLB071A is used as the starting point, the JCL names are  not listed in the impact analysis output. Resolution: After analyzing the step open Define Candidate Selection Criteria. In the Relations with Starting Points, the IN relations must be ticked. A document is attached to this Knowledge Base article showing how to retrieve the JCLs as impacted candidates. Incident #2491125 Attachments Impact Analysis - JCL retrieved as a result.doc Old KB# 33720#EnterpriseAnalyzer#JCL#SupportTip#ImpactAnalysis

Created On:  19 March 2011 Problem: We are connecting to a MQ Series version7 queue manager running on Windows by defining environment variable MQSERVER=SYSTEM.DEF.SVRCONN/TCP/10.52.253.18(1428).  We use a standard MQ explorer to put a test message (i.e in ASCII).The problem is that when we do a MQGET and set MQGMO_CONVERT option in MQGMO structure the message is not converted to EBCDIC.  The message is put with the MQSTR option and when forwarding the same message to same program running in a real IBM CICS mainframe environment, the message is converted correctly.Is there anyway to configure it such that it can work correctly in MFE CICS environment as well? Resolution: Set TX_MQ_CONVERT Windows environment variable to an appropriate codeset value.  For example for US, set TX_MQ_CONVERT=037. Incident #2494697 Old KB# 33721#EnterpriseDeveloper#MFDS#MQGMO_CONVERT#TX_MQ_CONVERT

Created On:  15 March 2011 Problem: What do I need to do to setup IMS Connect in my Enterprise Server region? Resolution: Add a Listener on the required port. It should be named IMS Connect and the checkbox set to CUSTOM. The custom text should be set to "mfcs-ims-connect". If tracing is required then add in the following to the configuration information of the IMS Connect Listener: [Trace]trace=yesdata=yesdata-trace=allall-max=0file= Old KB# 33687#MFDS#EnterpriseDeveloper

Created On:  15 March 2011 Problem: Where is the information stored about the variables I "monitor" when animating my program. I have an MFE user/PC where the monitored variables never re-appear in the watch list. Resolution: These variables are stored in memory and when you close the project they are written to a file called projectname.DBS. This is an XML file which can be read with a normal text editor.When the project is opened, this file is then read back into memory. Incident #2496586 Old KB# 33689#MFDS#EnterpriseDeveloper

Created On:  24 March 2011 Problem: How can I determine what Micro Focus COBOL products (if any) are installed on a UNIX or Linux machine? Resolution: Use a "find" command to search the machine for files named "cobver".Subordinate to each directory where a Micro Focus COBOL product was installed on a Linux/UNIX machine, will be a file named "cobver".  It is a short text file containing version information -- "cobver" is short for "cobol version".  The file is present in all Micro Focus UNIX/Linux COBOL products, including:Object COBOL Developer's Suite (OCDS)Application Server for Object COBOL Developer's SuiteServer ExpressApplication Server for Server ExpressServer for COBOLStudio Enterprise Edition (i.e. Server Express 6.x)Server Enterprise Edition Visual COBOL Enterprise Developer COBOL ServerIt is possible more than one Micro Focus COBOL product is installed on a particular UNIX or Linux machine -- in that case, more than one "cobver" file will be discov

Created On:  11 February 2011 Problem: There is no uninstall script to remove the Enterprise Server, Server for SOA or Server for Enterprise Edition products on any of the UNIX platforms. Resolution: The Enterprise Server (ES) product was introduced/made available from v4.0 onwards.  Since then the name has changed to Server for SOA or Server for Enterprise Edition depending on the version purchased. -   Logon as the ES Admin user and stop any running Enterprise Servers either via the ES Admin webpage  or the "casstop" command:        casstop -rESNAME          or          casstop /rESNAME where ESNAME is the name of the enterprise server to stop.-   Cd to the $COBDIR/bin directory and run the command to stop the Enterprise Server License Manager (eslm):      ./eslmfgetpv k  &nb

Created On:  24 March 2011 Problem: How can I submit jobs to the Enterprise Server internal reader ? Resolution: The MicroFocus emulation of JES supports the IBM z/OS internal reader (INTRDR) for submitting batch jobs from within CICS or from a batch job, with certain restrictions. N.B. All job streams must be composed of records less than or equal to 80 characters. Records longer than this will either be ignored, cause JCL errors, or result in the job stream "disappearing" (i.e. it will not appear in any of the ES/JES spool queues).1) Submitting one or more batch jobs from another batch jobAny program can submit a batch job to the internal reader by writing the job stream to a DDname that is defined as follows: //ANYNAME DD SYSOUT=(*,INTRDR)The job stream can be defined in an external file, as records within the program that writes to the internal reader, or as instream data, provided that an alternative end-of-data delimiter is used, for example://INSTREAM JOB 'Inter

Created On:  17 March 2011 Problem: When you submit JCL to the internal reader using a REXX, the system changes the data set name by prefixing it with the username even though you have placed it within quotes.For example ,when running this REXX exec/* REXX  */ . . . ADDRESS TSO 'SUBMIT "USER.PROJECT.CNTL(JCL001)" ' .Exit 0you will receive this error message  ..DATA SET mfuser."USER.PROJECT.CNTL(JCL001)" NOT FOUND Resolution: You are using the single and double quote marks in the wrong place.This syntax will run as expected../* REXX  */ADDRESS TSO "SUBMIT 'USER.PROJECT.CNTL(JCL001)'  "Exit 0Note that the whole SUBMIT comand is double-quoted, the dataset name is single quoted inside the command.  Although ,in REXX, single and double quotation marks can normally be freely interchanged; fully qualified data set names must be in single quotes.  The failing syntax may appear to have run successfully on a mainframe, but t

Created On:  23 March 2011 Problem: We are getting this error when trying to stop a region on the command-line using casstop /r:CASCF0050S casgate: Console daemon not active, RC = 0015 Resolution: This usually means that the region was started via the Enterprise Server Admin screen instead of via the command-line.  Since the region was started using a different userid (system userid) than that of the current user, casstop /r will not work.To allow casstop to terminate a region started either by another user or via Enterprise Server Admin, you must use the /l option instead of /r (casstop /l).Example:casstop /lREGIONThis way casstop will perform a lookup on the region name and will be able to communicate with the directory server to initiate a shutdown properly. Old KB# 33747 Additionally: Note that only one user can start a region. Ensure that no other users are attempting to start the same region from other locations, such as via the Enterprise Server Administration