Implementing CICS Applications

This chapter contains the following sections:

•	Presentation of the z/OS Simple Application

•	Verifying the CICS Application Installation

•	Presentation of Simple Application on COBOL-IT / BDB

•	Implementing Synchronous CICS Transactions With a Limited Number of Parallel Instances

•	Implementing Asynchronous CICS Non-Delayed Transactions

•	Implementing Asynchronous CICS Delayed Transactions

•	Implementing CICS Application Using Temporary Storage (TS) Queues

•	Implementing Distributed Program Link (DPL)

•	Implementing CICS Common Work Area (CWA)

•	Implementing a CICS Transaction Work Area (TWA)

•	Implementing CICS Transaction Trigger Monitor (ARTCKTI)

•	Implementing Transferring CICS Regions

•	Implementing Intersystem Communication

•	Implementing Submitting JCL Online

•	Implementing Printing CICS Runtime Applications Data

•	Implementing Invoking Web Services from CICS Applications

•	CICS Runtime Logs

•	The CICS Runtime Server Logs

•	Disabling and Enabling Programs

•	CICS Runtime C Program Support

Presentation of the z/OS Simple Application

Introduction

This application was initially developed on a z/OS platform implementing COBOL programs used in batch and CICS contexts with VSAM and QSAM files and DB2 tables.

Data was unloaded from z/OS and converted and reloaded on a UNIX platform using Oracle Tuxedo Application Rehosting Workbench.

The language components were converted or translated from z/OS to UNIX by Oracle Tuxedo Application Rehosting Workbench.

These components use two major Oracle Tuxedo Application components, Batch Runtime and CICS Runtime, to emulate the technical centralized features of their original z/OS environment. Here, we will focus on the particular case of the CICS Runtime implementing COBOL Programs using CICS statements and DB2 statements.

This Simple Application manages the customers of a company thru a set of classical functions like creation, modification and deletion.

Description of the CICS Simple Application Components

All of the CICS components are declared with the same name, in the z/OS CICS CSD File. All of the resource declarations are made inside a z/OS CICS GROUP named PJ01TERM. This group is declared in the z/OS CICS LIST PJ01LIST used by CICS at start up to be automatically installed.

Mapsets

 

Table 4‑1 Simple Application Mapsets

Name	Description
RSSAM00	Customer maintenance entry menu
RSSAM01	Customer data inquiry screen
RSSAM02	Customer data maintenance screen (create, update and delete customer)
RSSAM03	Customer list screen

Programs

 

Table 4‑2 Simple Application Programs

Name	Description
RSSAT000	Customer maintenance entry program
RSSAT001	Customer data inquiry program
RSSAT002	Customer data maintenance program (new customer, update and delete customer)
RSSAT003	Customer list program

Transactions Codes

 

Table 4‑3 Simple Application Transactions Codes

Name	Description
SA00	Main entry transaction code (program RSSAT000)
SA01	Customer inquiry (program RSSAT001)
SA02	Customer maintenance (program RSSAT002)
SA03	Customer list (program RSSAT003)

VSAM File

 

Table 4‑4 Simple Application VSAM File

DDName	DataSetName	Description
ODCSF0	PJ01AAA.SS.VSAM.CUSTOMER	VSAM Main Customer File

Configuring a Standard CICS Application With CICS Runtime

The first example uses the CICS Simple File-to-Oracle application which uses only a z/OS VSAM File converted into a UNIX Oracle Table.

In our example, all of the UNIX components resulting from platform migration are stored in the trf directory.

The COBOL programs and BMS mapsets should be compiled and available as executable modules in the respective directories ${HOME}/trf/cobexe and ${HOME}/trf/MAP_TCP.

CICS Simple File-to-Oracle Application UNIX Components

COBOL Program Files

The ${HOME}/trf/cobexe directory contains the Simple Application CICS executable programs:

•	${HOME}/trf/cobexe/RSSAT000.gnt

•	${HOME}/trf/cobexe/RSSAT001.gnt

•	${HOME}/trf/cobexe/RSSAT002.gnt

•	${HOME}/trf/cobexe/RSSAT003.gnt

The Mapset Files

The ${HOME}/trf/MAP_TCP directory contains the Simple Application Data z/OS BMS mapsets compiled:

•	${HOME}/trf/MAP_TCP/RSSAM00.mpdef

•	${HOME}/trf/MAP_TCP/RSSAM01.mpdef

•	${HOME}/trf/MAP_TCP/RSSAM02.mpdef

•	${HOME}/trf/MAP_TCP/RSSAM03.mpdef

CICS Runtime Configuration

For a standard application, in addition to the initial settings, the following CICS resources in the same Group must be implemented:

•	Basic CICS transactions (synchronous and simultaneous).

•	COBOL Programs without SQL statements, CICS TS queues.

•

Mapsets.

•	VSAM file (logical name and associated data accessors).

To configure these resources:

1.	Declare these resources in their respective CICS Runtime Resource Configuration File.

2.	Configure the CICS Runtime Tuxedo Servers Groups and Servers to manage these resources. See Reference for a full description of which configuration files are used with each server.

Declaring CICS Resources to the CICS Runtime

Each resource is declared in the file corresponding to its type (program, transaction …). Each resource defined in a resource file must belong to a group.

In the following examples using the CICS Simple File-to- Oracle Application, we will use the CICS Runtime Group name SIMPAPP and all our *.desc files will be located in the ${home}/trf/config/resources directory.

Note:

In these configuration files, each line beginning with a "#" is considered as a comment and is not processed by CICS Runtime

Declaring CICS Transactions Codes

These declarations are made by filling the transactions.desc file for each transaction you have to implement.

For each transaction you have to declare in a csv format

1.	The name of the transaction (mandatory).

2.	The CICS Runtime Group name (mandatory).

3.	A brief description of the transaction (optional, at least one blank).

4.	The name of the program started by this transaction (mandatory).

In the File-to-Oracle Simple Application example, we have to declare four transactions: SA00, SA01, SA02 and SA03 in the SIMPAPP Group, starting the corresponding COBOL programs RSSAT000, RSSAT001, RSSAT002 and RSSAT003.

Once filled, the transactions.desc file looks like this:

Listing 4‑1 Simple Application transactions.desc File

#Transaction Name ;Group Name ; Description ;Program Name

SA00;SIMPAPP; Home Menu Screen of the Simple Application;RSSAT000

SA01;SIMPAPP; Customer Detailed Information Screen of the Simple Application;RSSAT001

SA02;SIMPAPP; Customer Maintenance Screen of the Simple Application;RSSAT002

SA03;SIMPAPP; Customer List of the Simple Application;RSSAT003

 

Declaring a CICS COBOL Program

All the programs used by the transactions previously declared, directly or indirectly through EXEC CICS statements like LINK, XCTL, START … must be declared in the same Group.

These declarations are made in the programs.desc file for each program to implement.

For each program you have to declare in a csv format:

1.	The name of the program (mandatory)

2.	The CICS Runtime Group name (mandatory)

3.	A brief description of the program (optional, at least one blank)

4.	The language in which the program is written COBOL (default)

In our Simple Application example, the only programs needed are RSSAT000, RSSAT001, RSSAT002 and RSSAT003 which are all coded in the COBOL language

Once filled, the programs.desc file looks like this:

Listing 4‑2 Simple Application programs.desc File

#PROGRAM;GROUP;DESCRIPTION;LANGUAGE;

RSSAT000;SIMPAPP; Home Menu Program of the Simple Application ;COBOL

RSSAT001;SIMPAPP; Customer Detailed Information Program of the Simple Application ;COBOL

RSSAT002;SIMPAPP; Customer Maintenance Program of the Simple Application

RSSAT003;SIMPAPP; Customer List of the Simple Application ;COBOL

 

Note:

Nothing is declared in the language field of RSSAT002, meaning that the LANGUAGE of this program is COBOL by default.

Declaring CICS Mapsets

To converse with end-users thru 3270 terminals or emulators, declare to CICS Runtime all of the physical mapsets (*.mpdef file) used in the COBOL programs previously defined thru the specific EXEC CICS statements described above in this document.

These declarations are made by filling the mapsets.desc file for each mapset you have to implement.

The input format of each of your mapset definitions must respect the following format description:

1.	On the first free physical line, type the [mapset] keyword.

2.	On the next line, enter the keyword name= followed by the name of your mapsets.

3.	On the next line, enter the keyword filename= followed by the physical path of your physical mapsets (.mpdef file).

In our Simple Application example, the mapsets used in our COBOL programs are RSSAM00, RSSAM01, RSSAM02 and RSSAM03.

Once filled, the mapsets.desc file looks like this:

Listing 4‑3 Simple Application mapsets.desc File

[mapset]

name=ABANNER

filename=<KIXDIR>/sysmap/abanner.mpdef [mapset]

name=RSSAM00

filename=<HOME>/demo/MAP_TCP/RSSAM00.mpdef

[mapset]

name=RSSAM01

filename=<HOME>/demo/MAP_TCP/RSSAM01.mpdef

[mapset]

name=RSSAM02

filename=<HOME>/demo/MAP_TCP/RSSAM02.mpdef

[mapset]

name=RSSAM03

filename=<HOME>/demo/MAP_TCP/RSSAM03.mpdef

 

Note:

The mapsets.desc file does not accept UNIX variables, so a fully expanded path must be provided in this file.

•	<KIXDIR>: must be replaced by the value of the ${KIXDIR} variable of the ~/.profile.

•	<HOME>: must be replaced by the value of the ${HOME} variable of the ~/.profile.

Declaring ISAM Files Resulting From a z/OS VSAM File Conversion

Previously, before declaring one or more files to CICS Runtime, all the physical components, files, accessor programs, COBOL Copybooks etc. must have been generated by the Oracle Tuxedo Application Rehosting Workbench Data components.

Among all the components built or converted by the Oracle Tuxedo Application Rehosting Workbench Data components, only accessor programs on converted VSAM files are used by CICS Runtime. The reason is that, once migrated, no file can be directly accessed. The file can only be accessed indirectly through an accessor program dedicated to the management of this file (one and only one accessor program per source file).

The Simple Application uses only the CUSTOMER Oracle table, resulting from the Oracle Tuxedo Application Rehosting Workbench Data Conversion of the z/OS VSAM KSDS file PJ01AAA.SS.VSAM.CUSTOMER.

So, for our File-to-Oracle application example, we have only one accessor, RM_ ODCSF0 (RM for Relational Module), to declare to CICS Runtime.

Note:

ODCSF0 represents the logical name previously defined in CICS that pointed to the physical file name PJ01AAA.SS.VSAM.CUSTOMER. Consequently, it is also the only file name known from the CICS COBOL program to access this file by EXEC CICS statements.

To Declare the ISAM Migrated Files:

1.	Modify the Tuxedo envfile adding a new variable, if not already present, describing all the VSAM/ISAM files used in the programs previously defined.

For our Simple Application example the following line must be entered, (for simplicity, we have located the file in the same place as the ubbconfig, envfile and tuxconfig files but this is not mandatory.

DD_VSAMFILE=${HOME}/trf/config/tux/desc.vsam

2.	If the file does not exist, physically create the desc.vsam file at the indicated location.

3.	Modify the desc.vsam file by adding a new line describing the different information fields used by the accessor in a "csv" format for each accessor/file used.

For our Simple Application example, the following line is entered:

Listing 4‑4 Simple Application ISAM File Declaration

#DDname;Accessor;DSNOrganization;Format;MaxRecordLength;KeyStart;KeyLength

ODCSF0;ASG_ ODCSF0;I;F;266;1;6

 

Where:

ODCSF0

Is the Data Description Name (logical name) used in the EXEC CICS Statements.

RM_ODCSF0

Is the name of the accessor program managing the access to the Oracle table resulting from the data conversion of the former VSAM File .

I

The Data Set Name organization is indexed

F

Fixed, all the records have the same fixed length format.

266

Maximum record length.

1

Key position in the file (1 means first column or first character).

6

Key length.

Modifying the CICS Runtime Tuxedo Servers

To manage CICS application transactions, in addition to the servers previously defined:

1.	Implement the CICS Runtime Tuxedo Server ARTSTRN.

This server manages only basic CICS Runtime transactions, those that are the most often used: synchronous (not delayed) and simultaneous (not only one at a time).

2.	Indicate to CICS Runtime to start only the transactions belonging to the SIMPAPP CICS Runtime Group name.

The following example of a *SERVERS section of the Tuxedo ubbconfig file shows the configuration of a ARTSTRN server.

Listing 4‑5 Simple Application CICS Runtime Server Tuxedo Configuration

*SERVERS

…

ARTSTRN SRVGRP=GRP02

SRVID=20

CONV=Y

MIN=1 MAX=1 RQADDR=QKIX110 REPLYQ=Y

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_strn -e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -s KIXR -l SIMPAPP "

…

 

Where

*SERVERS

Tuxedo ubbconfig Keyword indicating a Server Section definition.

SRVGRP

Is the Tuxedo Group Name to which ARTSTRN belongs.

SRVID

Is the identifier of a Tuxedo Server of ARTSTRN.

CONV=Y

Indicates that this server operates in a conversational mode.

MIN=1 and MAX=1

Indicates that only one instance of this server must be run.

REPLYQ=Y

Indicates that this server will respond.

RQADDR=QCNX015

Name of the Tuxedo queue used for the responses.

CLOPT

Is a quoted text string passed to the server containing its parameters.

-o

Indicates the file used for the standard output messages of the server.

-e

Indicates the file used for the error output messages of the server.

-r

Is a Tuxedo parameter used to provide statistical reports.

-s KIXR

Indicates the CICS Runtime name where the KIXR transaction is run.

-l SIMAPP

Indicates that only the transaction of the SIMAPP group are to be selected.

Modifying the CICS Runtime Tuxedo Servers Groups

To be started, the ARTSTRN server must be defined in a Tuxedo Server Group previously defined (and not commented) in the ubbconfig file.

In our example, ARTSTRN belong to the Tuxedo Server Group GRP02 (SRVGRP=GRP02).

Listing 4‑6 Simple Application CICS Runtime Tuxedo Servers Groups Example:

*GROUPS

…

GRP02

GRPNO=12

ENVFILE="/home2/work9/demo/config/tux/envfile"

TMSNAME="TMS_ORA"

…

 

Where

*GROUPS

Tuxedo ubbconfig Keyword indicating a Server Section Group section definition.

GRPNO=

Tuxedo Group.

ENVFILE=

Path of the Tuxedo envfile.

TMSNAME=

Name of the Tuxedo Transaction Manager Server executable.

Verifying the CICS Application Installation

Using the Tuxedo tmadmin psr Commands

Enter the Tuxedo tmadmin psr command to check that all of the CICS Runtime required servers (ARTTCPL, ARTCNX, and ARTSTRN) are running and that their messages conform to the Tuxedo documentation and this document.

Listing 4‑7 tmadmin psr Simple Application Installation Check

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- ---------------

BBL 200933 KIXR 0 2 100 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 2 100 ( IDLE )

ARTSTRN QKIX110 GRP02 20 6 300 ( IDLE )

 

> quit

#

 

Using the Tuxedo tmadmin psc Commands

Another possible check can be made by entering the Tuxedo tmadmin psc command to display all the different Tuxedo Services running.

In addition to the CICS Runtime System transactions/services (CSGM, CESN, CESF …), you can now see the transaction codes of your CICS Runtime application SA00, SA01, SA02 and SA03

Listing 4‑8 tmadmin psc Simple Application Installation Check

# tmadmin

...

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 1 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 1 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 3 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 3 AVAIL

 

> quit

#

Using the CICS Runtime Application

Before using the CICS application, you have to populate the ISAM files accessed by your application.Then, access CICS Runtime with a 3270 Terminal or Emulator, with a UNIX x3270 command. It should be:

# x3270 ${HOSTNAME}:${TCPNETADDR}

Where:

${HOSTNAME}

Is the System UNIX variable containing the name of the UNIX machine on which you are running CICS Runtime.

${TCPNETADDR}

Is the port number for your UNIX 3270 emulator set up by your Tuxedo Administrator at installation time in the ubbconfig file.

1.	You will receive the Good Morning Message.

2.	Clear it by pressing the Clear key of your 3270 emulator keypad.

3.	Type the main transaction code SA00 (of your CICS Runtime application) in the top left corner:

Figure 4‑1 Simple Application transaction Code Entry

4.	The main menu of the application is displayed:

Figure 4‑2 Simple Application Main Menu

5.	Navigate through the screens of the application to check that they are displayed without errors.

Presentation of Simple Application on COBOL-IT / BDB

Based on BDB with XA protocol, the CICS COBOL programs compiled by COBOL IT can access the indexed ISAM files which are converted from Mainframe VSAM files through the ART Workbench. The following sections describes the configurations should be done in ART CICS Runtime to enable this application.

Configuring ubbconfig File in CICS Runtime

Add the MRM parameter in the group entry of *GROUPS and *RMS section in Tuxedo ubbconfig file. See the following example:

Listing 4‑9 Adding MRM Parameter in ubbconfig File Example

*GROUPS

GRP02

GRPNO=12

MRM=Y

*RMS

MRMG_RM1

SRVGRP=GRP02

RMID=15

TMSNAME="TMS_BDB"

OPENINFO="BERKELEY-DB:/home2/work10/data"

 

Where:

*GROUPS

Tuxedo ubbconfig keyword indicating the definitions of Servers Groups.

GRPNO

Indicates the Tuxedo Group.

MRM= Y

Indicates that this server group can support multiple resource managers.

*RMS

Tuxedo ubbconfig keyword indicating the definitions of resource managers.

MRMG_RM1

Indicates the logical name of RMS entry.

SRVGRP

Indicates the name of the group associated with this RM.

RMID

Indicates the unique ID number of this RM in the group. ID number must be between 1 and 31.

TMSNAME

Indicates the name of the transaction manager server associated with the group specified by SRVGRP.

OPENINFO

Indicates the resource manager dependent information needed when opening resource manager for the associated group.

Building BDB TMS Server

To build the BDB TMS server, add the first three lines of the following text in $TUXDIR/udataobj/RM:

# BDB

BDB_HOME=/opt/cobol-it-64-bdb

BERKELEY-DB:db_xa_switch:-L/opt/cobol-it-64-bdb/lib -ldb-5

 

Build BDB TMS

buildtms -v -r BERKELEY-DB -o $APPDIR/TMS_BDB

Exporting Variables Before Booting Up ART Servers

Export the following variables explicitly before booting up the ART servers:

•	DD_VSAMFILE

export DD_RBDB02=${DATA}/MTWART.ES.SFI.RCIBDB02.BDB0122

RBDB02 is the logical file name.

•	COB_ENABLE_XA

export COB_ENABLE_XA=1

Implementing Synchronous CICS Transactions With a Limited Number of Parallel Instances

In some particular cases, the number of transactions bearing the same transaction code running simultaneously has to be limited, for performance constraints for example.

On z/OS, this limit cannot be defined in the transaction resource itself but is defined in a distinct resource named TRANCLASS (transaction class) that contains a specific MAXACTIVE parameter describing the maximum number of concurrent instances of the same transaction.

To link a transaction to a transaction class, to inherit its parameters, especially the MAXACTIVE parameter, the z/OS CICS transaction resource has a TRANCLASS field containing the name of the TRANCLASS resource.

This instance management is performed differently on UNIX with CICS Runtime. The maximum number of transactions running concurrently is defined by the number of servers offering the same transaction. This maximum number and the minimum number are indicated respectively in the MAX and MIN parameters of the ARTSTRN definition in the *SERVERS section of the Tuxedo file ubbconfig.

It means that the maxactive parameter is not taken in account to manage these limits except in the following very particular case:

The Special Case of Transaction Classes With MAXACTIVE=1

The MAXACTIVE=1 is really an exception in this management because it indicates that no concurrent transaction belonging to these kind of transaction classes can be run simultaneously.

To manage this very particular case of sequential transactions, a Tuxedo CICS Runtime feature must be configured

Modification of the ubbconfig File for Sequential Transactions

All of the transactions linked to transactions classes with a MAXACTIVE superior or equal to 2 are managed by the CICS Runtime Tuxedo Server ARTSTRN and do not required modifying anything else. For the transactions with a MAXACTIVE parameter set to 1, an CICS Runtime Tuxedo Server named ARTSTR1 is dedicated to their specific management.

To activate this server, modify the ubbconfig file to add this server in the *SERVERS section:

Listing 4‑10 Adding a ARTSTR1 Server to ubbconfig

*SERVERS

…

ARTSTR1 SRVGRP=GRP02

SRVID=200

CONV=Y

MIN=1 MAX=1

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_str1 -e /home2/work9/demo/Logs/TUX/sysout/stderr_str1 -r -- -s KIXR -l SIMPAPP"

…

 

Where:

*SERVERS

Tuxedo ubbconfig Keyword indicating a Server Section definition.

SRVGRP

Is the Tuxedo Group Name to which ARTSTR1 belongs.

SRVID

Is the identifier of a ARTSTR1 Tuxedo Server.

CONV=Y

Indicates that this server operates in a conversational mode.

MIN=1 and MAX=1

Are mandatory and indicate that only one instance of this server must run.

CLOPT

Is a quoted text string passed to the server containing the parameters:

-o

Indicates the file used for the standard output messages of the server.

-e

Indicates the file used for the error output messages of the server.

-r

Is a Tuxedo parameter used to produce statistical reports.

-s

KIXR indicates the CICS Runtime name where the KIXR transaction is run.

-l SIMAPP

Indicates that only the transaction of the SIMAPP group are to be selected.

Note:

All of the CICS Runtime Transaction Servers (ARTSTRN, ARTSTR1, ARTATRN and ARTATR1) share the same CICS Runtime Transaction Group Servers, no modifications are required to the ubbconfig Server Group Section (*GROUPS).

Modifying the tranclasses.desc File

For ART CICS, concurrent transactions do not really need to be bound to transactions classes with MAXACTIVE parameters superior or equal to two because parallelism is the default behavior.

For sequential transactions, it is mandatory because it is the only way to declare these transactions to CICS Runtime. Declare specific transaction classes defined with a MAXACTIVE=1 parameter. Like the other CICS Runtime resources, this one must belong to an CICS Runtime Group name. For each TRANCLASS, declare in a csv format:

1.	The name of the transaction class (mandatory)

2.	The CICS Runtime Group name (mandatory)

3.	A brief description of the transaction class (optional, at least one blank)

4.	The maximum number of the same transaction to RUN (MAXACTIVE).

Note:

The MAXACTIVE parameter should be understood like a binary switch:

•	MAXACTIVE=1 <=> Sequential transaction class (mandatory).

•	MAXACTIVE>1 (all the values are at this step equivalent) <=> Concurrent transaction (optional).

Examples:

TRCLASS1;SIMPAPP ; Tranclass with maxactive set to 1; 1

TRCLASS2;SIMPAPP ; Tranclass with maxactive set to 2; 2

TRCLAS10;SIMPAPP ; Tranclass with maxactive set to 10; 10

The first transclass TRCLASS1 has is maxactive parameter equal to 1, indicating that all the transaction belonging to this transclass must be managed sequentially by the ARTSTR1.

The two last tranclasses, TRCLASS2 and TRCLASS10, are in fact similar because their maxactive parameters are superior to 1 indicating that the transactions belonging to these tranclasses can run concurrently managed by the ARTSTRN server.

Note:

These two last definitions are optional. Their absence has the same meaning.

Modifying the transactions.desc File

In addition to the first four mandatory fields of this csv format file (Transaction name, Group name, Description, Program name), you must add a twelfth field: TRANCLASS (Transaction Class name).

The TRANCLASS field must be separated from the Program field by eight semicolon characters (';') with at least one blank between each of them.

In our example, let us suppose that the CICS Runtime Simple Application must have the following MAXACTIVE limits:

•	SA00: MAXACTIVE=0

•	SA01: MAXACTIVE=1

•	SA02: MAXACTIVE=2

•	SA03: MAXACTIVE=10

Then these transactions must be linked to the following tranclasses that we have previously defined:

•	SA00: none

•	SA01: TRCLASS1

•	SA02: TRCLASS2

•	SA03: TRCLAS10

Once modified, the transactions.desc file will look like this:

Listing 4‑11 Example transactions.desc File

#Transaction Name ;Group Name ; Description ;Program Name

SA00;SIMPAPP; Home Menu Screen of the Simple Application;RSSAT000

SA01;SIMPAPP; Customer Detailed Information Screen of the Simple ; Application;RSSAT001; ; ; ; ; ; ; ;TRCLASS1

SA02;SIMPAPP; Customer Maintenance Screen of the Simple Application;RSSAT002; ; ; ; ; ; ; ; TRCLASS2

SA03;SIMPAPP; Customer List of the Simple Application;RSSAT003; ; ; ; ; ; ; ; TRCLASS10

 

Notes:

•	No modification is made to SA00 meaning that no transaction class is associated with this transaction code. It means that this transaction is not associated with a MAXACTIVE=1 parameter and so is not sequential.

•	SA02 and SA03 are associated to transaction classes, respectively TRCLASS2 and TRCLASS10, defined with MAXACTIVE >= 2. Knowing that these transactions are not required, the result would be the exactly the same if SA02 and SA03 were defined like SA00 without transaction classes.

•	SA01, which can run sequentially, is the only one where the transaction class field is mandatory. Verify that its associated transaction class, TRCLASS1, is really defined with a MAXACTIVE=1.

Checking the ARTSTR1 Configuration

Using the Tuxedo tmadmin psr Commands

The ARTSTR1, is shown below:

Listing 4‑12 Checking the ARTSTR1 Server with the tmadmin psr Commands

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- ---------------

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL 200933 KIXR 0 3 150 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

 

> quit

#

Using the Tuxedo tmadmin psc Commands

No new service or transaction should appear.

In our example where ARTSTRN was the only server running, we can see that nothing changed when ARTSTR1 is also activated.

Listing 4‑13 Checking the ARTSTRN Server with the tmadmin psc Commands

# tmadmin

...

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

 

> quit

#

Implementing Asynchronous CICS Non-Delayed Transactions

These transactions are launched by specifics CICS EXEC CICS START TRANSID requests coded in the CICS programs that are not using DELAY or TIME parameters to delay their execution.

If at least one of your programs contains this kind of statement, install, and activate some new features of CICS Runtime Tuxedo Severs without changing any other settings.

Modifying the Tuxedo ubbconfig File to Manage Asynchronous Transactions

The file is modified in the same manner as for the ARTSTRN and the ARTSTR1 servers, except the "s" (synchronous) character used to prefix the name of these servers should be replaced by the "a" (asynchronous) character.

Using Parallel Asynchronous Transactions

To use parallel asynchronous transactions, with a MAXACTIVE parameter strictly superior to one, the dedicated server is the ARTATRN. Please refer to the section describing the installation of the ARTSTRN server to install the atrn_server.

To check your settings you can use also the tmadmin psr and psc commands.

For the Simple Application example we can see that:

•	The psr command shows that a new server is running ARTATRN.

•	The psc command shows that five new services are running, one is dedicated to the asynchronous transaction while each synchronous transaction (SA00 to SA03) is duplicated (ASYNC_SA00 to ASYNC_SA03) to allow them to run in an asynchronous mode.

Listing 4‑14 tmadmin Commands Showing Parallel Asynchronous Transactions

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL           200933 KIXR 0 4 200 ( IDLE )

ARTTCPL       00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTATRN QKIXATR GRP02 30 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

ASYNC_QUEUE ASYNC_QUEUE ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA03 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA02 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA01 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA00 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

 

> quit

{deimos:work9}-/home2/work9/demo/config/tux#{deimos:work9}-/home2/work9/demo/config/tux#

 

Using Non-Parallel Asynchronous Transactions

To use non-parallel asynchronous transactions, with a MAXACTIVE parameter exactly equal to one, the dedicated server is ARTATR1.

Please refer to the section describing the reasons and the installation of the ARTSTR1 server to install the ARTSTR1 server.

To check your setting, you can use also the Tuxedo tmadmin psr and psc commands

For the Simple Application example we can see that:

•	The psr command shows that a new server is running ARTATR1.

•	The psc command shows that no new services are running.

Listing 4‑15 tmadmin Commands Showing non-parallel Asynchronous Transactions

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL 200933 KIXR 0 4 200 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

 

> quit

#

 

Implementing Asynchronous CICS Delayed Transactions

ART CICS Runtime supports two methods for implementing asynchronous CICS delayed transactions launched using EXEC CICS START TRANSID requests:

•	Implementing Asynchronous Transactions With ARTSRM Server (Recommended)

•	Implementing Asynchronous Transactions With /Q

Implementing Asynchronous Transactions With ARTSRM Server

On z/OS, there are some time-related CICS START API options can be used to start a transaction at a specified time or after a specified interval, such as AT, TIME, AFTER, and INTERVAL. ART CICS Runtime provides a server, ARTSRM, for implementing these options. For more information, refer to ARTSRM Configuration in Oracle Tuxedo Application Runtime for CICS Reference Guide.

To activate this server, configure ARTSRM in the *SERVERS section in the UBBCONFIG file. You can configure a set of ARTSRM servers only if they are in the same group for each CICS region. Following is an example.

Listing 4‑16 Example of Configuring ARTSRM in UBBCONFIG

*SERVERS

…

ARTSRM

SRVGRP=ARTGRP

SRVID=500

RESTART=Y

MAXGEN=5

GRACE=3600

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_srm -e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -s KIXR -l SIMPAPP"

 

Note:

Although implementing asynchronous transactions with /Q is still supported, when the START TRANID/CANCEL command is invoked, the request is submitted to the TRANCTL_[SYSID] service advertised by ARTSRM firstly. If the call gets the TPNOENT fail code, then use /Q to redispatch the request.

Implementing Asynchronous Transactions With /Q

Asynchronous transactions are launched when ASYNC_QSPACE for EXEC START is set with option INTERVAL or PROTECT.

In this case, the transaction request is deposited into a Oracle Tuxedo /Q Queue, and when the time is ready, the transaction will be automatically invoked.

For this feature to be available, these components must be configured:

1.	An Oracle Tuxedo /Q Queue Space named ASYNC_QSPACE

2.	An Oracle Tuxedo /Q Queue named ASYNC_QUEUE in ASYNC_QSPACE.

3.	The TMQUEUE and TMQFORWARD servers dedicated to these asynchronous transactions.

Creating the Tuxedo /Q

CICS Runtime provides a UNIX script that creates all the Tuxedo /Q components: mkqmconfig.sh.

1.	Before using the script, define and export in your UNIX ~./.profile file:

•	The QMCONFIG variable QMCONFIG- containing the full directory path that stores the Tuxedo /Q Queue Space ASYNC_QSPACE.

•	The KIX_QSPACE_IPCKEY variable - containing the IPC Key for the Queue Space.

Examples of ~/.profile variables and values:

export QMCONFIG=${HOME}/trf/config/tux/kixqspace

export KIX_QSPACE_IPCKEY=200955

2.	Execute mkqmconfig.sh from the command line to create the Tuxedo /Q features.

Modifying the Tuxedo ubbconfig File to Manage the Tuxedo /Q Queue

1.	The GQUEUE Server Group must be added to the ubbconfig file in the *GROUP section.

Listing 4‑17 Simple Application Tuxedo Queue ubbconfig Example

*GROUPS

…

# /Q

GQUEUE GRPNO=1000

TMSNAME=TMS_QM TMSCOUNT=2

OPENINFO="TUXEDO/QM:/home2/work9/demo/config/tux/kixqspace:ASYNC_QSPACE"

…

 

Where:

*GROUPS

Tuxedo ubbconfig Keyword indicating definitions of Servers Groups.

GRPNO=

Tuxedo Group.

TMSCOUNT=

Number of Tuxedo Transaction Manager Servers.

TMSNAME

Name of the Tuxedo Transaction Manager Server executable.

OPENINFO=

Indicates to the Tuxedo /Q Transaction Manager QM, the QSPACE name to manage and its UNIX absolute path.

2.	Then, two servers, TMQUEUE and TMQFORWARD, must be added to the ubbconfig file in the *SERVERS section.

Listing 4‑18 Simple Application ubbconfig TMQUEUE and TMQFORWARD Example

*SERVERS

…

# /Q

TMQUEUE SRVGRP=GQUEUE

SRVID=1010

GRACE=0 RESTART=Y CONV=N MAXGEN=10

CLOPT="-s ASYNC_QSPACE:TMQUEUE -- "

TMQFORWARD

SRVGRP=GQUEUE

SRVID=1020

GRACE=0 RESTART=Y CONV=N MAXGEN=10

CLOPT="-- -n -i 2 -q ASYNC_QUEUE"

…

 

Where:

*SERVERS

Tuxedo ubbconfig Keyword indicating a Server Section definition.

SRVGRP

Is the Tuxedo Group Name which the server belongs to.

SRVID

Is the identifier of a Tuxedo Server.

MAXGEN=10

Specifies that the process can have up to 10 server restarts.

GRACE=0

Means there is no limit interval to contain the number of server restarts.

CONV=N

Indicates that this server operates in a non-conversational mode.

CLOPT

Is a quoted text string passed to the server containing its parameters.

Using the tmadmin psr and psc commands check that four new servers and two new services are running:

Listing 4‑19 Simple Application TMQUEUE and TMQFORWARD tmadmin Example

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL           200933 KIXR 0 4 200 ( IDLE )

ARTTCPL       00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

TMS_QM        GQUEUE_TMS GQUEUE 30001 0 0 ( IDLE )

TMS_QM        GQUEUE_TMS GQUEUE 30002 0 0 ( IDLE )

TMQUEUE       01000.01010 GQUEUE 1010 0 0 ( IDLE )

TMQFORWARD    01000.01020 GQUEUE 1020 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTATRN QKIXATR GRP02 30 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

TMS TMS TMS_QM GQUEUE 30001 KIXR 0 AVAIL

TMS TMS TMS_QM GQUEUE 30002 KIXR 0 AVAIL

ASYNC_QSPACE TMQUEUE TMQUEUE GQUEUE 1010 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

ASYNC_QUEUE ASYNC_QUEUE ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA03 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA02 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA01 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA00 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

 

> quit

#

 

Implementing CICS Application Using Temporary Storage (TS) Queues

These transactions use CICS programs containing EXEC CICS requests relative to CICS Temporary Storage Queues.

The statements used are EXEC CICS WRITEQ TS … END-EXEC, EXEC CICS READQ TS … END-EXEC, EXEC CICS DELETEQ TS … END-EXEC.

If at least one of your programs contains one of these statements, install and activate the new features of CICS Runtime without changing your other settings.

To manage TS Queues, activate the ARTTSQ CICS Runtime Tuxedo Server.

•	To activate this server, add this server to the *SERVERS section of the Tuxedo ubbconfig file:

Listing 4‑20 Activating the ARTTSQ in the ubbconfig File

*SERVERS

…

ARTTSQ SRVGRP=GRP02

SRVID=40

MIN=1 MAX=1

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_tsq -e /home2/work9/demo/Logs/TUX/sysout/stderr_tsq -r -- -s KIXR -l SIMPAPP"

…

 

Where:

*SERVERS

Tuxedo ubbconfig Keyword indicating a Server Section definition.

SRVGRP

Is the Tuxedo Group Name to which ARTTSQ belongs.

SRVID

Is the identifier of a Tuxedo Server of ARTTSQ.

MIN=1 and MAX=1

Indicates that only one instance of this server must be run.

CLOPT

Is a quoted text string passed to the server containing its parameters:

-o

Indicates the following file is used for the standard output messages of the server.

-e

Indicates the following file is used for the error output messages of the servers.

-r

Is a Tuxedo parameter used to have statistical reports.

-s KIXR

Indicates the CICS Runtime name where the transaction runs is KIXR.

-l SIMAPP

Indicates that only the components of the SIMAPP group are to be selected at start up.

Use the Tuxedo tmadmin psr and psc commands to check that the server is running and that six new services are published:

Listing 4‑21 Checking ARTTSQ Server and Services are Running

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL           200933 KIXR 0 3 150 ( IDLE )

ARTTCPL       00001.00101 TCP00 101 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTTSQ 00012.00040 GRP02 40 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

TSM00004_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00003_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00002_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00001_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00000_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSQUEUE tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

 

> quit

{deimos:work9}-/home2/work9/demo/config/tux#

 

Implementing Unrecoverable TS Queues

For unrecoverable TS Queues, no integrity is guaranteed by CICS Runtime concerning their content. For example, if an abend occurs at any point during a CICS transaction, work done on this TS is not rolled-back to the last consistency point.

TS Queues are stored in a sequential file in a dedicated directory defined in the KIX_TS_DIR UNIX environment variable. This variable is defined and then exported from the ~/.profile UNIX System File:

KIX_TS_DIR=${HOME}/trf/KIXTSDIR

Modify the Tuxedo ubbconfig file to activate the new ARTTSQ server dedicated to their management.

Implementing Recoverable TS Queues

For these TS Queues, CICS Runtime guarantees the integrity of their content. For example, if an abend occurs at any point during a CICS transaction, they are rolled-back to the last consistency point, if all is in order, their content is committed to become a new consistency point. These TS Queues are stored in Oracle Tables to benefit from the RDBMS integrity management.

Concerning the TS Queue, there is an enhanced behavior for reading a recoverable TS Queue.

On source CICS z/OS, CICS enqueuing is not invoked for READQ TS commands, thereby making it possible for one task to read a temporary storage queue record while another is updating the same record. To avoid this, use explicit enqueuing on the temporary storage queues so that concurrent executing tasks can read and change queues with the same temporary storage identifier.

This behavior also allows one transaction to see or read a record freshly written in a recoverable TS Queue, even before it is committed, and after its rollback.

On target we don't have this limitation, but in particular:

•	A reading transaction is not able to see a record that is just added and not yet committed.

•	A reading transaction is not able to see a modification to the record that is not yet committed.

To Use Recoverable TS Queues

To use recoverable TS Queues you need to define an Oracle Table to contain the TS Queues. CICS Runtime provides a UNIX script to create all these tables: crtstable_Oracle.

1.	Before using the script define and export from your UNIX ~./.profile file

•	The ORA_USER variable containing the user ID used to connect to Oracle.

•	The ORA_PASSWD variable containing the associated password.

Examples of ~/.profile variables and values:

export ORA_USER="Oracle_User_1"

export ORA_PASSWD="Oracle_Pswd_1"

2.	Once the variables have been set, execute the crtstable_Oracle script.

3.	Then, modify the Tuxedo ubbconfig file to modify the Server Group used by ARTTSQ to establish the connection to Oracle in the *GROUPS section.

Listing 4‑22 Example of the *GROUP Section of the Tuxedo ubbconfig File Concerning the Derver Group GRP02 used by the ARTTSQ Server.

*GROUPS

…

GRP02

GRPNO=12

ENVFILE="/home2/work9/demo/config/tux/envfile"

TMSNAME="TMS_ORA" OPENINFO="Oracle_XA:Oracle_XA+Acc=P/work9/work9+SesTm=600+LogDir=/home2/work9/demo/Logs/TUX/xa+DbgFl=0x20"

…

 

Where:

*GROUPS

Tuxedo ubbconfig Keyword indicating definitions of Servers Groups.

GRPNO=

Tuxedo Group number.

TMSNAME=

Name of the Tuxedo Transaction Manager Server executable.

OPENINFO=

Parameters send to the Oracle_XA Manager.

4.	Use the Tuxedo psr and psc commands to check that Oracle is available; three new servers and three new services should be indicated:

Listing 4‑23 Simple Application Check For Recoverable TS Queues

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL           200933 KIXR 0 4 200 ( IDLE )

ARTTCPL       00001.00101 TCP00 101 0 0 ( IDLE )

TMS_ORA       GRP02_TMS GRP02 30001 0 0 ( IDLE )

TMS_ORA       GRP02_TMS GRP02 30002 0 0 ( IDLE )

TMS_ORA       GRP02_TMS GRP02 30003 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTTSQ 00012.00040 GRP02 40 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

TMS TMS TMS_ORA GRP02 30001 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30002 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30003 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

TSM00004_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00003_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00002_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00001_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00000_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSQUEUE tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

 

> quit

#

 

Managing TD Queue Intrapartititions

Presentation of the Mechanism on Source Platform

Transient Data Control

The CICS transient data control facility provides a generalized queuing facility. Data can be queued (stored) for subsequent internal or external processing. Selected data, specified in the application program can be routed to or from predefined symbolic transient data queues: either intrapartition or extrapartition.

Transient data queues are intrapartition if they are associated with a facility allocated to the CICS region and extrapartition if the data is directed to a destination that is external to the CICS region. Transient data queues must be defined and installed before the first reference by an application program.

You can:

•	Write data to a transient data queue (WRITEQ TD command).

•	Read data from a transient data queue (READQ TD command).

•	Delete an intrapartition transient data queue (DELETEQ TD command).

Note:

In this document we concentrate exclusively on intrapartition TD queues.

Intrapartition Transient Data Queues

Intrapartition refers to data on direct-access storage devices for use with one or more programs running as separate tasks. Data directed to or from these internal queues is referred to as intrapartition data; it must consist of variable-length records.

When data is written to the queue by a user task, the queue can be used subsequently as input data by other tasks within the CICS region. All access is sequential, governed by read and write pointers. Once a record has been read, it cannot be read subsequently by another task. Intrapartition data may ultimately be transmitted upon request to the terminal or retrieved sequentially from the output dataset.

Typical uses of intrapartition data include:

•	Message switching.

•	Broadcasting.

•	Database access.

•	Routing of output to several terminals (for example, for order distribution).

•	Queuing of data (for example, for assignment of order numbers or priority by arrival).

•	Data collection (for example, for batched input from 2780 Data Transmission Terminals)

There are three types of intrapartition transient data queues:

Non-recoverable

Non-recoverable intrapartition transient data queues are recovered only on a warm start of CICS. If a unit of work (UOW) updates a non-recoverable intrapartition queue and subsequently backs out the updates, the updates made to the queue are not backed out.

Physically recoverable

Physically recoverable intrapartition transient data queues are recovered on warm and emergency restarts. If a UOW updates a physically recoverable intrapartition queue and subsequently backs out the updates, the updates made to the queue are not backed out.

Logically recoverable

Logically recoverable intrapartition transient data queues are recovered on warm and emergency restarts. If a UOW updates a logically recoverable intrapartition queue and subsequently backs out the changes it has made, the changes made to the queue are also backed out. On a warm or an emergency restart, the committed state of a logically recoverable intrapartition queue is recovered. In-flight UOWs are ignored.

Automatic Transaction Initiation (ATI)

For intrapartition queues, CICS provides the option of automatic transaction initiation (ATI).

A basis for ATI is established by the system programmer by specifying a non-zero trigger level for a particular intrapartition destination. When the number of entries (created by WRITEQ TD commands issued by one or more programs) in the queue reaches the specified trigger level, a transaction specified in the definition of the queue is automatically initiated. Control is passed to a program that processes the data in the queue; the program must issue repetitive READQ TD commands to deplete the queue.

When the queue has been emptied, a new ATI cycle begins. That is, a new task is scheduled for initiation when the specified trigger level is again reached, whether or not execution of the earlier task has ended. The exact point at which a new ATI cycle begins depends on whether or not the queue is defined as logically recoverable. If the queue is defined with a RECOVSTATUS of No or Physical, the new ATI cycle begins when the queue is read to QZERO. But if the queue is defined with a recoverability attribute of Logical, the new ATI cycle begins only after the task terminates after having read the queue to QZERO.

If an automatically initiated task does not empty the queue, access to the queue is not inhibited. The task may be normally or abnormally ended before the queue is emptied (that is, before a QZERO condition occurs in response to a READQ TD command). If the contents of the queue are to be sent to a terminal, and the previous task completed normally, the fact that QZERO has not been reached means that trigger processing has not been reset and the same task is reinitiated. A subsequent WRITEQ TD command does not trigger a new task if trigger processing has not been reset.

Presentation of the Mechanism on Target Platform

Tuxedo /Q

Tuxedo /Q offers a robust and versatile queuing system with the same capabilities as TD queues and more.

Queues can be defined as recoverable or not, and triggering with a few different options is also available. The management of errors is much more sophisticated, and will simplify error management in case of ATI transaction failures on target.

Architecture Design

 

Table 4‑5 Source to Target Mapping

Source Element	Target Correspondence
TD Queue intrapartition	Tuxedo /Q Queue.
Associated transaction (TRANID)	Associated transaction offered by an ATR server.
Trigger level	Trigger level.
Recoverability: No|Physical|Logical	Similar levels available as on target, but with different configuration principles.

The CICS verbs READQ TD, WRITEQ TD and DELETEQ TD (applied to intrapartition queues), now read, write or delete from a Tuxedo /Q queue. (tpenqueue and tpdequeue) in terms of tuxedo vocabulary.

If the Queue is logically recoverable, these actions are done in the current UOW, else they are done atomically, independently of the current UOW.

For information, inside CICS Runtime, this is done by adding the TPNOTRAN flag to operations on non-logically recoverable queues.

Triggering

In case of triggering, like in native CICS, a transaction will be automatically triggered, this transaction having to read the corresponding queue and process accordingly the messages.

In CICS Runtime these asynchronous transactions are offered and processed by a dedicated server type ARTATR, with either of its two variants ARTATR1 and ARTATRN.

These servers process all asynchronous transactions, more precisely, transactions submitted by START TRANSID, or by automatic Transaction Invocation related to td queue intrapartition.

In this case a specific CICS Runtime client, TDI_TRIGGER, is used to launch the corresponding asynchronous transaction, when the trigger level is reached.

Runtime CICS Configuration of TD Queue Intrapartition

CICS Runtime Resource Declaration

Every CICS-like resource in CICS Runtime, is declared using a dedicated configuration file stored in directory ${KIXCONFIG}.

TD Queue extrapartition and TD Queue intrapartition resource declaration share very few arguments, and are semantically very different objects, even if using the same API for read and write operations.

This is the reason why, in CICS Runtime, we have separated TD Queue extrapartition resource configuration and TD Queue intrapartition resource configuration into two different resource files.

Intrapartition queues are declared in the file tdqintra.desc, described in Oracle Tuxedo Application Runtime for CICS Reference Guide.

The important attributes are:

TDQUEUE(name)

The queue name, exactly identical to the queue name in the source configuration, This name must be the same as the name of the queue inthe Tuxedo /Q configuration.

RECOVSTATUS(status)

Only the status NO or LOGICAL, are accepted, the difference between the two modes impacts the treatment of WRITEQ TD and READQ TD, more precisely LOGICAL making them part of the current UOW, while NO makes them atomic operations independent of the current UOW.

The difference between NO or PHYSICAL, is not defined in the resource configuration file but will be implemented using native tuxedo /Q configuration parameters, mapping to persistent /Q or non persistent.

TRANSID and TRIGGERLEVEL

In the current release are documentary only in tdqintra.desc, it is their value in /Q configuration which is taken in account.

QSPACENAME

New argument needed for /Q: defining into which QSPACE the current /Q is stored. This argument is mandatory and must match the QSPACENAME into which the actual /Q queue is physically stored.

/Q Configuration for TD Queue Intrapartition in CICS Runtime

For detailed and accurate information on qmadmin and /Q configuration Using the ATMI /Q Component in the Tuxedo documentation.

The script mk_td_qm_config.sh distributed with CICS Runtime provides an example of qspace creation and then of queue creation and configuration into /Q, to be used for TD intrapartition queues.

This script uses three environment variables, which must be set according to your environment:

•	KIX_TD_QSPACE_DEVICE: must contain the filename of the physical file containing the /Q database for TD queues.

•	KIX_TD_QSPACE_NAME: contains the name of the logical QSPACE to create, which will contains the queues.

•	KIX_TD_QSPACE_IPCKEY: a specific key which must be unique on the machine for the IPC used by the instance of /Q.

The creation of the device (KIX_TD_QSPACE_DEVICE) and of the QSPACE are very standard, we will not detail them.

The interesting part is related to queue configuration.

A qopen QspaceName command, to open the qspace which will contain the queues must be made before the creation of any queue. The QspaceName must match the QSPACENAME in the resource declaration of these queue(s).

Below is an example of an interactive queue creation using qmadmin, where the questions asked by qmadmin are in normal font, while the entries typed in by the user are in bold.

Listing 4‑24 qopen Dialog

qopen TD_QSPACE

qcreate

Queue name: TEST

Queue order (priority, time, expiration, fifo, lifo): fifo

Out-of-ordering enqueuing (top, msgid, [default=none]): none

Retries [default=0]: 5

Retry delay in seconds [default=0]: 0

High limit for queue capacity warning (b for bytes used, B for blocks used,

% for percent used, m for messages [default=100%]): 5m

Reset (low) limit for queue capacity warning [default=0%]: 0m

Queue capacity command: "TDI_TRIGGER -t S049"

 

In a script an exact equivalent to this manual entry would be:

Listing 4‑25 qopen Script

qopen TD_QSPACE

qcreate TEST fifo none 3 0 5m 0m "TDI_TRIGGER -t S049"

 

qopen Parameters

TD_QSPACE

The QspaceName must match the QSPACENAME in the resource declaration of these queue(s).

Queue name

The name of the queue must match exactly the name provided in the resource declaration.

Queue order

The default dequeuing order when reading the queue, the setting corresponding to TD intra native behavior is: fifo.

Out-of-ordering enqueuing

Not meaningful unless some application is using native /Q interface to write into these queue; for Runtime CICS only usage to set it to is: none

Retries

Defines the number of times a message will be put back on the queue in case of abort of the UOW having read this queue, to avoid resubmitting again and again an ATI transaction which fails because of a bad message, set this number to a reasonable number.

When this number is reached, or at the first abort if you set it to zero, the message will be removed from this queue and put onto the error queue for further analysis.

Retry delay in seconds

If retries is not null, defines a delay before putting a record back on its queue, in case of rollback, the recommended value with Runtime CICS is the default value 0.

High limit for queue capacity warning

This is the much more flexible equivalent of the trigger level of TD queues. For a setting compatible with TD queues, set it to the trigger level and express it in number of messages. For example: 0m to suspend triggering, or 5m for a trigger level of 5 messages in the queue.

Reset (low) limit for queue capacity warning

This is the down level to be reached before resetting the trigger for the upper limit, for compatibility with TD queue behavior, it should be set to 0, (QZERO) which is the reset value for TD queues in CICS.

Queue capacity command:

This is the command to be launched when the trigger level is reached, in CICS Runtime it should be set to: TDI_TRIGGER -t TRID. Where TRID is the Transaction identifier of the transaction to trigger which should match the TRANSID of the resource configuration.

Tip:

ATR servers when processing an ATI, know whether the transaction reached QZERO or not, and whether it was a success or a rollback. So if QZERO is not reached, they resubmit the transaction in the same manner as on the source platform. But now it is the number of retries which will replace the ATIFACILITY parameter and will govern the fact that a rollbacked TD queue record will be resubmitted or not. It is a progress compared with the source is that now the administrator can decide the number of resubmissions, and get the faulty messages on an error queue.

Activating the ARTTDQ in the Tuxedo ubbconfig File

To enable TDQ motoring, ARTTDQ server should be activated.

 

 

 

 

 

 

 

 

 

Listing 4‑26 Activating the ARTTDQ in the ubbconfig File

*SERVERS

…

ARTTDQ SRVGRP=GRP02

SRVID=40

MIN=1 MAX=1

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_tdq -e /home2/work9/demo/Logs/TUX/sysout/stderr_tdq -r -- -s KIXR -l SIMPAPP"

…

 

Where:

*SERVERS

Tuxedo ubbconfig Keyword indicating the definition of Server Section.

SRVGRP

The Tuxedo Group Name to which ARTTDQ belongs.

SRVID

The identifier of a Tuxedo Server of ARTTDQ.

MIN=1 and MAX=1

Indicates that only one instance of this server must be run.

CLOPT

A quoted text string passed to the server containing its parameters:

-o

Indicates the following file is used for the standard output messages of the server.

-e

Indicates the following file is used for the error output messages of the servers.

-r

Is a Tuxedo parameter used to have statistical reports.

-s KIXR

Indicates the CICS Runtime name where the transaction runs is KIXR.

-l SIMAPP

Indicates that only the components of the SIMAPP group are to be selected at start up.

Implementing Distributed Program Link (DPL)

For several reasons, on z/OS, the Distributed Program Link function enables a local CICS program (the client program) to call another CICS program (the server program) in a remote CICS region via EXEC CICS LINK statements. CICS Runtime supports this feature used in multi-CICS architecture like MRO among migrated regions.

To Detect That DPL Is Needed

Unless you wish to use the DPL in a UNIX written application, check the technical specificities of the z/OS application

1.	Check on z/OS, using the CEDA system transaction, if at least one remote program is defined in the z/OS CICS CSD file. Such programs have some of their fields of the REMOTE ATRIBUTES section filed:

Listing 4‑27 Checking for Remote Programs

DEF PROGR

OVERTYPE TO MODIFY CICS RELEASE = 0610

CEDA DEFine PROGram( )

PROGram ==>

Group ==>

DEscription ==>

....

REMOTE ATTRIBUTES

DYnamic ==> No No ! Yes

REMOTESystem ==> XXXX

REMOTEName ==> YYYYYYYY

Transid ==> ZZZZ

EXECUtionset ==> Dplsubset Fullapi ! Dplsubset

 

Where (CICS default values are underlined):

DYNAMIC(YES|NO)

The following parameters cannot be overridden in the CICS LINK API. This field is only relevant for DPL use when it is set to NO and the three following fields are empty.

REMOTESYSTEM(name)

Remote CICS region name. An empty field is not relevant with DYNAMIC(YES)

REMOTENAME(name)

Remote server program name. An empty field is not relevant with DYNAMIC(YES) because the default is the client program name (PROGram ==>).

TRANSID(name)

Remote mirror transaction. An empty field is not relevant with DYNAMIC(YES) because the default is the mirror system transaction CSMI.

EXECUTIONSET(FULLAPI|DPLSUBSET)

The DPL cannot use the full CICS API but only a subset. The DPLSUBSET parameter indicates explicit usage of a DPL subset of the CICS API, but note that this subset may also be sufficient to execute LINK in a non-DPL context without errors. On the other hand, this field may contain FULLAPI in a DPL context but does not ensure that no "Invalid Request errors" will follow if non-DPL API are used.

As described above, in some cases, the Remote Attributes declaration may not exist or can be incomplete. The reason is that these fields establish only some of the default values, some of the previous parameters in bold in the example are not provided in the EXEC CICS LINK API.

2.	Then check in the programs, inside the EXEC CICS LINK API:

•	If the names of the programs called in this order match the names of programs defined in the CSD with remote attributes partially or fully informed.

•	If these statement contain at least one of the optional remote parameters shown in italics in the following CICS LINK API (the others fields are not relevant for DPL).

Listing 4‑28 CICS LINK API For DPL

EXEC CICS LINK PROGRAM(…)

COMMAREA(…)

LENGTH(…)

DATALENGTH(…)

RETCODE(…)

SYSID(XXXX) : Remote CICS region name

SYNCONRETURN : Used for remote CICS syncpoint or rollback

TRANSID(XXXX) : Remote mirror transaction instead of the CSMI default

INPUTMSG(…)

INPUTMSGLEN(…)

END-EXEC

Modifying the Tuxedo ubbconfig File to Manage the DPL

If at least one of your programs use the DPL, install and activate the ARTDPL server without changing your other settings.

To activate this server, modify your ubbconfig file to add this server to the *SERVERS section of the Tuxedo ubbconfig file. This server belongs to the same Server Group as the Transactions Servers (ARTSTRN, ARTSTR1, ARTATRN, ARTATR1).

Listing 4‑29 ubbconfig File Example of a *SERVERS Section Describing the ARTDPL Server.

*SERVERS

…

ARTDPL SRVGRP=GRP02

SRVID=500

CONV=N

MIN=1 MAX=1 RQADDR=QKIXDPL REPLYQ=Y

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_dpl -e /home2/work9/demo/Logs/TUX/sysout/stderr_dpl -r -- -s KIXD -l SIMPAPP"

…

 

Where:

*SERVERS

Tuxedo ubbconfig Keyword indicating a Server Section definition.

SRVGRP

Is the Tuxedo Group Name to which ARTDPL belongs.

SRVID

Is the identifier of a Tuxedo Server of ARTDPL.

CONV=N

Indicates that this server operates in a non-conversational mode.

MIN=1 and MAX=1

Indicates that only one instance of this server must be run.

REPLYQ=Y

Indicates that this server will respond.

RQADDR=QKIXDPL

Name of the Tuxedo queue used for the responses.

CLOPT

Is a quoted text string passed to the server containing its parameters:

-o

Indicates the following file is used for the standard output messages of the server.

-e

Indicates the following file is used for the error output messages of the server.

-r

Is a Tuxedo parameter used to provide statistical reports.

-s KIXD

Indicates the CICS Runtime name where the KIXD transaction is run.

-l SIMAPP

Indicates that only the components of the SIMPDPL group are to be selected at start up.

Use the Tuxedo tmadmin psr and psc commands to check that this server is running and that no new service is published:

Listing 4‑30 tmadmin Commands to Check ARTDPL Server

# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- --------------

ARTDPL QKIXDPL GRP02 500 0 0 ( IDLE )

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL 200933 KIXR 0 5 250 ( IDLE )

TMS_QM GQUEUE_TMS GQUEUE 30001 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30001 0 0 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

TMS_QM GQUEUE_TMS GQUEUE 30002 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30002 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30003 0 0 ( IDLE )

TMQUEUE 01000.01010 GQUEUE 1010 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

TMQFORWARD 01000.01020 GQUEUE 1020 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTATRN QKIXATR GRP02 30 0 0 ( IDLE )

ARTTSQ 00012.00040 GRP02 40 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

TMS TMS TMS_QM GQUEUE 30001 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30001 KIXR 0 AVAIL

TMS TMS TMS_QM GQUEUE 30002 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30002 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30003 KIXR 0 AVAIL

ASYNC_QSPACE TMQUEUE TMQUEUE GQUEUE 1010 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA02 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

ASYNC_QUEUE ASYNC_QUEUE ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA03 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA02 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA01 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA00 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

TSQUEUE tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

 

> quit

#

 

Declaring Remote Programs in CICS Runtime

To allow an application to use distributed programs called in EXEC CICS LINK statements, these programs must be declared to CICS Runtime.

1.	To declare REMOTE programs which can only use the DPL Subset of the CICS API:

•	In the programs.desc file, set REMOTESYSTEM (the 7th field of the csv format dataset), to remote SYSID name (KIXD in sample of Listing 4‑29).

The default is local (empty field), meaning that local programs are declared because they can use the FULL CICS API.

In our Simple Application example, if we suppose that RSSAT000, RSSAT001 are remote and RSSAT002 and RSSAT003 are local, then the programs.desc file is set to:

Listing 4‑31 Simple Application programs.desc Configuration of Remote Programs

#PROGRAM;GROUP;DESCRIPTION;LANGUAGE;EXECKEY;STATUS;REMOTESYSTEM;REMOTENAME

RSSAT000;SIMPAPP;Home Menu Program of Simple Application;COBOL; ;ENABLE;KIXD

RSSAT001;SIMPAPP;Customer Detailed Inf Program of Simple Application;COBOL; ;ENABLE;KIXD

RSSAT002;SIMPAPP;Customer Maintenance Program of the Simple Application;COBOL; ;ENABLE

RSSAT003;SIMPAPP;Customer List of the Simple Application;COBOL; ;ENABLE

 

2.	Shutdown and reboot Tuxedo.

3.	Using the Tuxedo tmadmin psr and psc commands, check that new services for DPL programs are published and managed by ARTDPL: KIXD_RSSAT0001 and KIXD_RSSAT0003.

Note:

To avoid problems with homonyms, these distributed services have their names composed of the Tuxedo DOMAINID defined in the ubbconfig and the name of the program they manage.

Listing 4‑32 Using tmadmin Commands to Check DPL Services

{deimos:work9}-/home2/work9/demo/Logs/TUX/sysout# tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- ---------------

ARTDPL QKIXDPL GRP02 500 0 0 ( IDLE )

ARTATR1 00012.00300 GRP02 300 0 0 ( IDLE )

ARTSTR1 00012.00200 GRP02 200 0 0 ( IDLE )

BBL 200933 KIXR 0 5 250 ( IDLE )

TMS_QM GQUEUE_TMS GQUEUE 30001 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30001 0 0 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

TMS_QM GQUEUE_TMS GQUEUE 30002 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30002 0 0 ( IDLE )

TMS_ORA GRP02_TMS GRP02 30003 0 0 ( IDLE )

TMQUEUE 01000.01010 GQUEUE 1010 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

TMQFORWARD 01000.01020 GQUEUE 1020 0 0 ( IDLE )

ARTSTRN QKIX110 GRP02 20 0 0 ( IDLE )

ARTATRN QKIXATR GRP02 30 0 0 ( IDLE )

ARTTSQ 00012.00040 GRP02 40 0 0 ( IDLE )

 

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

KIXD_RSSAT0+ dplsvc ARTDPL GRP02 500 KIXR 0 AVAIL

KIXD_RSSAT0+ dplsvc ARTDPL GRP02 500 KIXR 0 AVAIL

TMS TMS TMS_QM GQUEUE 30001 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30001 KIXR 0 AVAIL

TMS TMS TMS_QM GQUEUE 30002 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30002 KIXR 0 AVAIL

TMS TMS TMS_ORA GRP02 30003 KIXR 0 AVAIL

ASYNC_QSPACE TMQUEUE TMQUEUE GQUEUE 1010 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

SA03 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA01 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

SA00 kixsvc ARTSTRN GRP02 20 KIXR 0 AVAIL

ASYNC_QUEUE ASYNC_QUEUE ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA03 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA01 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

ASYNC_SA00 atrsvc ARTATRN GRP02 30 KIXR 0 AVAIL

TSM00004_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00003_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00002_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00001_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSM00000_TSQ tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

TSQUEUE tsqsvc ARTTSQ GRP02 40 KIXR 0 AVAIL

 

> quit

# .

 

To see full details on the truncated values displayed, you can use the Tuxedo verbose command.

To reduce the scope of the services listed to only those managed by ARTDPL (SRVID=500), use the Tuxedo psc command followed with the -i srvid parameter to restrict the display to a particular server id.

In our example, the srvid of the ARTDPL server is 500 as displayed just above.

Listing 4‑33 Using tmadmin Commands to Check Specific DPL Service in Verbose Mode

# tmadmin

...

 

> verbose

Verbose now on.

 

> psc -i 500

Service Name: KIXD_RSSAT003

Service Type: USER

Routine Name: dplsvc

Prog Name: /home2/work9/KIXEDO/bin/ARTDPL

Queue Name: QKIXDPL

Process ID: 1327244, Machine ID: KIXR

Group ID: GRP02, Server ID: 500

Current Load: 50

Current Priority: 50

Current Trantime: 30

Current Blocktime: 0

Current BUFTYPECONV: 0

Requests Done: 0

Current status: AVAILABLE

 

Service Name: KIXD_RSSAT001

Service Type: USER

Routine Name: dplsvc

Prog Name: /home2/work9/KIXEDO/bin/ARTDPL

Queue Name: QKIXDPL

Process ID: 1327244, Machine ID: KIXR

Group ID: GRP02, Server ID: 500

Current Load: 50

Current Priority: 50

Current Trantime: 30

Current Blocktime: 0

Current BUFTYPECONV: 0

Requests Done: 0

Current status: AVAILABLE

 

> quit

#

Implementing CICS Common Work Area (CWA)

On z/OS, the CWA is a common storage area defined in memory for a CICS region that programs can use to save and exchange data between themselves as long as this CICS region is running.

This area is addressed thru a pointer delivered by the CICS statement EXEC CICS ADDRESS CWA. If you find this CICS statement in your application, you have to implement this feature in CICS Runtime.

Listing 4‑34 COBOL Example of CWA Usage

LINKAGE SECTION.

01 COMMON-WORK-AREA.

03 APPL-1-ID PIC X(4).

03 APPL-1-PTR USAGE IS POINTER.

03 APPL-2-ID PIC X(4).

03 APPL-2-PTR USAGE IS POINTER.

PROCEDURE DIVISION.

. . .

END-EXEC.

* Set up addressability to the CWA

EXEC CICS ADDRESS

CWA(ADDRESS OF COMMON-WORK-AREA)

END-EXEC.

 

After the CICS ADDRESS CWA, the address of the COBOL group named COMMON-WORK-AREA is set to the address of the CWA allocated by CICS, meaning that COMMON-WORK-AREA maps and refines this memory area. The total amount of this shared memory is fixed and defined at CICS start up.

To Replicate CICS ADDRESS CWA Functionality in CICS Runtime

1.

Contact your z/OS CICS Administrator to know the size of memory implemented. (For your information this value is defined with the parameter WRKAREA of the DFHSIT. The default value is 512 bytes and the size can vary from 0 to 3584 bytes). Another way is to calculate the biggest size of the data record contained in the programs addressing the CWA.

2.	Modify your ~/.profile UNIX system file to export a new CICS Runtime variable, KIX_CWA_SIZE, and set it to the value found in the WRKAREA of the DFHSIT. If this variable is not declared, note that the default value is 0 and the authorized interval from 0 to 32760 bytes.

Example:

KIX_CWA_SIZE=512

3.	Modify your ~/.profile UNIX system file to export a new CICS Runtime variable, KIX_CWA_IPCKEY, and valorize it to a Unix IPC key to define the cross memory segment used as CWA.

Example:

KIX_CWA_ IPCKEY=200944

4.	Restart Tuxedo to take all these changes into account.

Implementing a CICS Transaction Work Area (TWA)

On z/OS, the TWA is a common storage area defined in memory for a CICS region that programs can use to save and exchange data between themselves during the execution time of one CICS transaction. In other words, this TWA can only be accessed by the programs participating in the transaction. This area is addressed thru a pointer delivered by the CICS statement EXEC CICS ADDRESS TWA. If you find an EXEC CICS ADDRESS TWA statement in your application, you have to implement this feature in CICS Runtime.

Listing 4‑35 A COBOL Example of Use of the TWA

LINKAGE SECTION.

01 TRANSACTION-WORK-AREA.

03 APPL-1-ID PIC X(4).

03 APPL-1-PTR USAGE IS POINTER.

03 APPL-2-ID PIC X(4).

03 APPL-2-PTR USAGE IS POINTER.

PROCEDURE DIVISION.

. . .

END-EXEC.

* Set up addressability to the TWA

EXEC CICS ADDRESS

TWA(ADDRESS OF TRANSACTION-WORK-AREA)

END-EXEC.

 

After the CICS ADDRESS TWA, the address of the COBOL group named TRANSACTION-WORK-AREA is set to the address of the TWA allocated by CICS, meaning that TRANSACTION -WORK-AREA maps and refines this memory area. The total amount of this shared memory is defined for each transaction in the z/OS CSD configuration file in the field TWasize.

The next screen shows the result of a z/OS CEDA system transaction where the TWasize parameter is set to 122 for the SA00 transaction code:

Figure 4‑3 z/OS ceda System Transaction Example

To replicate this functionality in CICS Runtime:

1.	Modify the CICS Runtime transactions.desc file to report the needed amount of TWA memory (TWasize>0).

2.	For each transaction using programs with CICS ADDRESS TWA statements, modify the transactions.desc file to declare its TWasize in the sixteenth field of this csv format file.

 

Table 4‑6 TWA Size Values Associated to Each Transaction Code of the Simple Application

Transaction	TWA Size
SA00	0
SA01	100
SA02	200
SA03	300

Listing 4‑36 Configuration of TWA in the transactions.desc File

#Transaction;Group;Description;Program; ; ; ; ; ; ;Status; ; ; ;Tranclass ;TWA Size

SA00;SIMPAPP;pg for simpapp;RSSAT000; ; ; ; ; ; ;ENABLED

SA01;SIMPAPP;pg for simpapp;RSSAT001; ; ; ; ; ; ;ENABLED; ; ; ; ;100

SA02;SIMPAPP;pg for simpapp;RSSAT002; ; ; ; ; ; ;ENABLED; ; ; ; ;200

SA03;SIMPAPP;pg for simpapp;RSSAT003; ; ; ; ; ; ;ENABLED; ; ; ; ;300

 

Note:

Nothing is indicated for the SA00 transaction that had a TWA size equal to zero.

3.	Restart the CICS Runtime Tuxedo servers, the modifications can be seen in the different stderr files of the servers involved in the transaction management (ARTSTRN, ARTSTR1, ARTATRN and ARTATR1)

Listing 4‑37 stderr_strn TWA Example

|---------------------------------|

| TRANSACTIONS loaded : < 4> |

|----------------------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

| | | | |C|C| |R|R| | |T| | | |

|TRAN| GROUP | PROGRAM |ALIA|M|O|PRI|E|E| STATUS |TASK |R| TRAN | TWA |MAX|

| | | | |D|N| |S|S| |DATA |A| CLASS | SIZ |ACT|

| | | | |S|F| |S|T| |KEY |C| | |IVE|

|----|----------|------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

|SA00|SIMPAPP |RSSAT000 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA01|SIMPAPP |RSSAT001 | |N|N|001|N|N|ENABLED |USER |Y| |00100|999|

|SA02|SIMPAPP |RSSAT002 | |N|N|001|N|N|ENABLED |USER |Y| |00200|999|

|SA03|SIMPAPP |RSSAT003 | |N|N|001|N|N|ENABLED |USER |Y| |00300|999|

 

Supporting TWA in ARTDPL

The programs within a transaction run by ARTDPL now can access TWA with following steps.

1.	Modify the CICS Runtime transactions.desc file to configure TWASize needed for ARTDPL transaction.

Listing 4‑38 Configuration of TWA for ARTDPL in the transactions.desc File

#Transaction;Group;Description;Program; ; ; ; ; ; ;Status; ; ; ;Tranclass ;TWA Size

CPMI;SIMPAPP;pg for simpapp;DFHMIRS; ; ; ; ; ; ;ENABLED; ; ; ; ;100

 

DFHMIRS is the internal mirror program in CICS that handles inbound function shipping. In CICS RT, this mirror program should be defined under the transaction used to run the linked program in the transaction resource file if TWA is used. In the list, ARTDPL runs remote linked program under transaction named CPMI and it has TWA size equal to 100.

Note:

Users should not name application program as DFHMIRS.

2.	Restart the CICS Runtime Tuxedo servers and the modifications can be seen in ARTDPL stdout file.

Listing 4‑39 stdout_dpl TWA Example

|---------------------------------|

| TRANSACTIONS loaded : < 1> |

|----------------------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

| | | | |C|C| |R|R| | |T| | | |

|TRAN| GROUP | PROGRAM |ALIA|M|O|PRI|E|E| STATUS |TASK |R| TRAN | TWA |MAX|

| | | | |D|N| |S|S| |DATA |A| CLASS | SIZ |ACT|

| | | | |S|F| |S|T| |KEY |C| | |IVE|

|----|----------|------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

|CPMI|SIMPAPP |DFHMIRS | |N|N|001|N|N|ENABLED |USER |Y| |00100|999|

|-----------------------------------------------------------------------------------------------------|

 

Implementing CICS Transaction Trigger Monitor (ARTCKTI)

The ART CICS Transaction Trigger Monitor (ARTCKTI) behaves the same as the CICS CKTI transaction. It listens on one or multiple WebSphere MQ initiation queues, retrieves trigger messages when a trigger event occurs, and then forwards the trigger messages to the target transaction.

Work Flow

ARTCKTI is a standalone Oracle Tuxedo server. The ARTCKTI server behaves as follows:

1.	Monitor one or multiple WebSphere MQ initiation queues.

One server instance can only monitor WebSphere MQ initiation queues within the same WebSphere MQ queue manager. The queues in different WebSphere MQ queue managers should be monitored by separate ARTCKTI server instances.

2.	When trigger message has arrived, the ARTCKTI server retrieves the message.

3.	Retrieve the transaction ID from the trigger message.

4.	Transfer the trigger message from structure MQTMC to MQTMC2.

Since MQTMC has many fields, it is always too complicated to send the structure as the parameter of EXEC CICS START call. MQTMC2 is used in CKTI to pass the structure as data to the START request for the trigger monitor.

5.	Invoke the target transaction, and send the MQTMC2 data.

Since CICS CKTI transaction starts the target transaction with asynchronized call (EXEC CICS START), the ARTCKTI server also starts the target transaction with asynchronized call (Tuxedo tpacall).

6.	User transaction retrieves the trigger message by CICS RETRIEVE, and performs operations on the WebSphere MQ application queue.

If the user transaction does not retrieve the message or the triggered transaction is not available, WebSphere MQ no longer sends trigger message in this condition. A new trigger message is issued until the WebSphere MQ initiation queue is reopened or a new trigger condition is met.

Figure 4‑4 illustrates the behavior.

Figure 4‑4 WebSphere MQ Trigger Condition

Command Configuration

ARTCKTI accepts the following parameters for the ubbconfig file.

•	-i trigger_interval: specifies the maximum time (in milliseconds) that the ARTCKTI server waits for a message to arrive at the WebSphere MQ initiation queue.

•	-s retry_interval: specifies the retry interval for ARTCKTI to reconnect to WebSphere MQ queue manager or reopen WebSphere MQ initiation queue upon failure.

•	-m queue_manager_name: specifies the name of the WebSphere MQ queue manager to be monitored.

•	-q queue1,queue2,……: specifies the name of the WebSphere MQ initiation queue to be monitored.

Implementing Transferring CICS Regions

In z/OS, ISSUE PASS command is used to transfer CICS regions without terminal reconnection; users can also implement data transference using LOGONMSG. When ISSUE PASS is invoked, the GMTRAN of destination region will be invoked by force.

ART CICS also supports the above scenario. Following configurations are required.

Configuring ARTSRM Server

It is required to configure ARTSRM. For more information, please refer to ARTSRM Configuration.

Configuring Environment Variables

It is required to set environment variable ISC_ENABLE to YES. For more information, please refer to ISC_ENABLE.

CICS Runtime Configuration Files Declaration

system.desc

system.desc defines system initialization parameters of CICS regions.

Listing 4‑40 Example for system.desc Configurations

[kixr]

APPLID=DBDCkixR

INITPARM=(ASINTP='Hello world')

[kixl]

APPLID=DBDCkixL

INITPARM=(ASINTP='Hello world')

GMTRAN=ISSS

LGNMSG=YES

 

In this example, two CICS regions are defined. SYSID are specified as kixr and kixl respectively. On one hand, kixl specifies GMTRAN=ISSS; when users log in DBDCkixL, transaction are invoked automatically. On the other hand, kixr doesn't specify GMTRAN; default CSGM is used. LGNMSG specified in kixl enables data transference function using ISSUE PASS in EXTRACT LOGONMSG. For more information about system.desc, please refer to System Configuration File.

transactions.desc and programs.desc

If GMTRAN (not other system transactions, such as CSGM, CESN, or CESF) is defined, transactions/programs should be configured in transactions.desc/programs.desc, and then loaded by ARTSTRN/ARTSTR1. For more information about transactions.desc/programs.desc, please refer to Transaction Configuration File and Programs Configuration File.

Listing 4‑41 Example for transactions.desc and programs.desc Configurations

transactions.desc:

ISSS;SIMPAPPB;pg for simpapp;ISSPASSS

programs.desc:

ISSPASSS;SIMPAPPB;pg for simpapp;COBOL; ;ENABLED

 

terminals.desc (Optional)

This configuration file defines terminal available to ART CICS; it is mandatory for using static LUNAME to logon ART CICS. For more information about terminals.desc, please refer to Terminal Configuration File.

Listing 4‑42 Example for terminals.desc Configurations

[terminal]

name=0001

netname=CICS0001

group=SIMPAPP

[terminal]

name=0002

netname=CICS0002

group=SIMPAPP

 

UBB Declaration

To implement transferring CICS regions, the following requirements should be met.

•	TMQUEUE should be configured for each CICS region.

•	ARTLOGN should be configured.

•	At least one ARTCNX should be configured for each CICS region.

•	DDR published by ARTCNX should be configured (an example is provided as below).

Listing 4‑43 Example for DDR Configurations

GRP00

GRPNO=10

ENVFILE="/home2/work9/demo/config/tux/envfile"

 

GRP01

GRPNO=11

ENVFILE="/home2/work9/demo/config/tux/envfile"

GRP02

GRPNO=12

ENVFILE="/home2/work9/demo/config/tux/envfile"

 

GQUEKIXR

GRPNO=1010

TMSNAME=TMS_QM TMSCOUNT=2

OPENINFO="TUXEDO/QM: /home2/work9/demo/sysfile/kixrqspace:DBDCkixR"

GQUEKIXL

GRPNO=1020

TMSNAME=TMS_QM TMSCOUNT=2

OPENINFO="TUXEDO/QM: /home2/work9/demo/sysfile/kixlqspace:DBDCkixL"

...

TMQUEUE

SRVGRP=GQUEKIXR

SRVID=1110

RESTART=Y GRACE=0 CONV=N MAXGEN=10

CLOPT="-s DBDCkixR:TMQUEUE -- "

TMQUEUE

SRVGRP=GQUEKIXL

SRVID=1210

RESTART=Y GRACE=0 CONV=N MAXGEN=10

CLOPT="-s DBDCkixL:TMQUEUE -- "

 

ARTCNX

SRVGRP=GRP01

SRVID=15

CONV=Y

MIN=1 MAX=1 RQADDR=QCNX015 REPLYQ=Y

CLOPT="-o /home2/work9/demo /LOGS/sysout/stdout_cnx_15 -e /home2/work9/demo /LOGS/sysout/stderr_cnx_15 -r -- -s KIXR -l SIMPAPP"

ARTCNX

SRVGRP=GRP02

SRVID=16

CONV=Y

MIN=1 MAX=1 RQADDR=QCNX016 REPLYQ=Y

CLOPT="-o /home2/work9/demo /sysout/stdout_cnx_16 -e /home2/work9/demo /LOGS/sysout/stderr_cnx_16 -r -- -s KIXL -l SIMPAPP"

ARTLOGN

SRVGRP=GRP00

SRVID=18

CONV=Y

MIN=1 MAX=1 RQADDR=QLGN018 REPLYQ=Y

CLOPT="-o /home2/work9/demo /LOGS/sysout/stdout_logn -e /home2/work9/demo /LOGS/sysout/stderr_logn -r --"

 

...

*SERVICES

DEFAULT: SVCTIMEOUT=0 TRANTIME=80

connect ROUTING=CICSISC

disconnect ROUTING=CICSISC

inquire ROUTING=CICSISC

update ROUTING=CICSISC

CSGM ROUTING=CICSISC

CESN ROUTING=CICSISC

CESF ROUTING=CICSISC

authfail ROUTING=CICSISC

 

*ROUTING

CICSISC FIELD=CX_APPLID RANGES="'DBDCKIXR':GRP01,'DBDCKIXL':GRP02,*:GRP01" BUFTYPE="FML32"

 

Notes:

DDR configuration in UBB is mandatory. DDR routes login request to ARTCNX in different CICS regions by FML field CX_APPLID.

The APPLID configured in ROUTING RANGES should be in upper case.

ART reserved FML FIELD ID from 8100 to 8191 for DDR.

Environment Variable Declaration

Set environment variable ISC_ENABLE=YES to transfer CICS regions.

Use Tuxedo tmadmin psr and tmadmin psc to check whether ARTLOGN starts successfully. ARTCNX and TMQUEUE are included in each region.

Listing 4‑44 Example for Environment Variable Declaration

/home2/work9/demo> tmadmin> tmadmin

...

 

> psr

Prog Name Queue Name Grp Name ID RqDone Load Done Current Service

--------- ---------- -------- -- ------ --------- ---------------

BBL 34790 KIXR 0 55 2750 ( IDLE )

TMS_QM GQUEKIXL_T+ GQUEKIXL 30001 0 0 ( IDLE )

TMS_QM GQUEKIXR_T+ GQUEKIXR 30001 0 0 ( IDLE )

ARTTCPL 00001.00101 TCP00 101 0 0 ( IDLE )

TMS_QM GQUEKIXL_T+ GQUEKIXL 30002 0 0 ( IDLE )

TMS_QM GQUEKIXR_T+ GQUEKIXR 30002 0 0 ( IDLE )

TMQUEUE 01020.01210 GQUEKIXL 1210 1 50 ( IDLE )

TMQUEUE 01010.01110 GQUEKIXR 1110 2 100 ( IDLE )

ARTADM 00011.00010 GRP01 10 0 0 ( IDLE )

ARTCNX QCNX015 GRP01 15 0 0 ( IDLE )

ARTCNX QCNX016 GRP02 16 0 0 ( IDLE )

ARTLOGN QLGN018 GRP00 18 0 0 ( IDLE )

ARTSTRN QKIX110 GRP12 20 0 0 ( IDLE )

...

> psc

Service Name Routine Name Prog Name Grp Name ID Machine # Done Status

------------ ------------ --------- -------- -- ------- ------ ------

DBDCkixL TMQUEUE TMQUEUE GQUEK+ 1210 KIXR 1 AVAIL

DBDCkixR TMQUEUE TMQUEUE GQUEK+ 1110 KIXR 2 AVAIL

disconnect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

update cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

inquire cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP01 15 KIXR 0 AVAIL

disconnect cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

connect cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

update cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

inquire cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

authfail cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

CESF cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

CESN cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

CSGM cnxsvc ARTCNX GRP02 16 KIXR 0 AVAIL

delsess lognsvc ARTLOGN GRP00 18 KIXR 0 AVAIL

gensess lognsvc ARTLOGN GRP00 18 KIXR 0 AVAIL

ART_LOGON lognsvc ARTLOGN GRP00 18 KIXR 0 AVAIL

ISSS strsvc ARTSTRN GRP12 25 KIXR 0 AVAIL

...

 

Logon ART CICS

After a successful boot up, users can connect to ART CICS, and then logon screen prompts out for users to specify the CICS region (APPLID) to logon.

Figure 4‑5 Logon Screen

Implementing Intersystem Communication

ART CICS Runtime supports implementing two z/OS intercommunication features:

•	Implementing Distributed Transaction Processing (DTP)

•	Implementing Asynchronous Processing

•	Implementing Synchronous Processing

Implementing Distributed Transaction Processing (DTP)

ART CICS supports DTP connections in multiple ART CICS regions through APPC mapped and LUTYPE6.1 protocol. On this view, COBOL applications using DTP (APPC/LUTYPE6.1) verbs can be deployed to ART CICS directly after being translated by Oracle Tuxedo Application Rehosting Workbench.

ART CICS also supports integration with Oracle TMA to enable DTP connections between ART CICS region and Mainframe CICS region through APPC. Following is a typical end-to-end user case.

Figure 4‑6 Typical End-to-End User Case

In this scenario, there are three ART CICS regions, KIXA, KIXB, and KIXC, among which KIXA and KIXB communicates with each other via APPC protocol, and KIXA and KIXC communicates with each other via LU61 protocol.

Besides, there is another CICS region called CICA on Mainframe, which communicates with either KIXA or KIXB via APPC protocol.

As shown in Figure 4‑6, the conversations occur in these regions are:

•	Conversations among ART CICS regions

•	APPC conversation issued from the terminal in KIXA to KIXB

•	KVA0 KIXB KVA5

•	KVA2 KIXB KVA5

•	LU61 conversation issued from the terminal in KIXA to KIXC

•	RV60 KIXC RV65

•	Conversations between ART CICS region and Mainframe CICS region through TMA

•	APPC outbound conversation issued from the terminal in KIXA to CICA

•	KVA0 CICA KVA5

•	KVA2 CICA KVA5

•	APPC inbound conversation issued from the terminal in CICA to KIXB

•	KVA0 CRM1 KVA5

•	KVA2 CRM1 KVA5

Note:

CRM1 is a connection from CICA to TMA LU.

Configurations

Following are configurations required for DTP connections to work in this scenario.

CICS Region Definitions in system.desc

Following CICS regions are defined in the system.desc configuration file:

•	Three ART CICS regions on open system:

•	KIXA: APPC/LU61 front end

•	KIXB: APPC back end

•	KIXC: LU61 back end

•	One CICS region on Mainframe:

•	CICA: APPC front / back end

For more information about system.desc, refer to System Configuration File.

Connections Definitions in connections.desc

Following connections are defined in the connection.desc configuration file:

•	Three connections are defined in KIXA:

•	KIXB: connects to KIXB, protocol is APPC

•	KIXC: connects to KIXC, protocol is LU61

•	CICA: connects to CICA, protocol is APPC

•	Two connections are defined in KIXB:

•	KIXA: connects to KIXA, protocol is APPC

•	CICA: connects to CICA, protocol is APPC

•	One connection is defined in KIXC:

•	KIXA: connects to KIXA, protocol is LU61

Note:

Connection definition for CICA is not presented on local because CICA is an external region.

For more information about connections.desc, refer to Connection Configuration File.

Programs Definitions in programs.desc

Following programs are defined in the programs.desc configuration file:

•	Two programs in KIXA:

•	COVSATMC: APPC client with view32

•	RVS61C: LU61 client

•	One program in KIXB:

•	COVSATMS: APPC server with view32

•	One program in KIXC:

•	RVS61S: LU61 server

For more information about programs.desc, refer to Programs Configuration File.

Transactions Definitions in transactions.desc

Following transactions are defined in the transactions.desc configuration file:

•	Three transactions in KIXA:

•	KVA0: APPC client on COVSATMC, sync level 0

•	KVA2: APPC client on COVSATMC, sync level 2

•	RV60: LU61 client on RVS61C

•	One transaction in KIXB:

•	KVA5: APPC server on COVSATMS

•	One transaction in KIXC:

•	RV65: LU61 server on RVS61S

For more information about transactions.desc, refer to Transaction Configuration File.

UBBCONFIG Configuration

Following are configured in the UBBCONFIG file:

•	One ARTSTRN server for KIXA, with three services defined in transactions.desc: KVA0, KVA2, and RV60

•	One ARTCTRN server for KIXB, with one service defined in transactions.desc: KIXB_KVA5

•	One ARTCTRN server for KIXC, with one service defined in transactions.desc: KIXC_RV65

•	GWSNAX gateway for TMA integration (for CICA)

For more information about ARTSTRN and ARTCTRN, refer to CICS Runtime Servers.

Note:

For ARTCTRN configurations in UBBCONFIG, it is required to specify CONV=Y.

DMCONFIG Configuration

Following are configured in the DMCONFIG file:

•	Remote domain definition for TMA integration (for CICA)

•	Import the CICA_KVA5 service from external

•	Export the KIXB_KVA5 service to external

Implementing Asynchronous Processing

On z/OS, asynchronous processing refers to a START command that starts a transaction on a remote system. ART CICS Runtime supports implementing this feature using the START command with a SYSID option.

The following sections describe the configuration tasks you need to perform.

Defining Regions in system.desc

Define your CICS regions in the system.desc configuration file. Following is an example that defines two regions, KIXR and KIXX, with respective application definitions, DBDCkixR and DBDCKIXX.

Listing 4‑45 Example of Defining Regions in system.desc

[KIXR]

APPLID=DBDCkixR

INITPARM=(ASINTP='Hello world')

[KIXX]

APPLID=DBDCKIXX

INITPARM=(ASINTP='Hello worldL')

 

Configuring ARTSRM Server

It is required to configure ARTSRM server. For more information, please refer to ARTSRM Configuration.

Modifying the UBBCONFIG File

It is required to configure ARTATRN servers in the UBBCONFIG file for each CICS region.

Suppose you have defined two regions, KIXR and KIXX, as shown in Listing 4‑45. Following is a configuration example.

Listing 4‑46 Example of Modifying UBBCONFIG

…

*SERVERS

…

ARTATRN

SRVGRP=GRP02

SRVID=30

CONV=N

MIN=1 MAX=1 RQADDR=QKIXATR REPLYQ=Y

CLOPT="-o /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stdout_atrn -e /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS /sysout/stderr_atrn -r -- -s KIXR -l SIMPAPP"

ARTSRM SRVGRP= GRPX SRVID=36 MIN=1 MAX=1 RQADDR= QKIXATR REPLYQ=Y CLOPT="-o /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stdout_srm -e /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stderr_srm -r -- -s KIXR -l SIMPAPP "

ARTATRN

SRVGRP=GRPX

SRVID=35

CONV=N

MIN=1 MAX=1 RQADDR=QKIXATRX REPLYQ=Y

CLOPT="-o /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stdout_atrn -e /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS /sysout/stderr_atrn -r -- -s KIXX -l SIMPAPP"

ARTSRM SRVGRP= GRPX SRVID=36 MIN=1 MAX=1 RQADDR= QKIXATRX REPLYQ=Y CLOPT="-o /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stdout_srm -e /u01/common/patches/yfli/KIX12110/test/CIT_ORA/strt/LOGS/sysout/stderr_srm -r -- -s KIXX -l SIMPAPP "

…

*SERVICES

 

Implementing Synchronous Processing

ART CICS Runtime supports invoking a synchronous transaction, which resides on a remote CICS system. To implement this feature, the following configurations are required.

Configuring Environment Variables

Set the ISC_ENABLE environment variable to YES to enable the synchronous processing feature.

Defining Regions in system.desc

Define your CICS regions in system.desc configuration file. Following is an example that defines two regions, KIXR and KIXX, with respective application definitions, DBDCKIXR and DBDCKIXX.

Listing 4‑47 Example of Defining Regions in system.desc

[KIXR]

APPLID=DBDCKIXR

[KIXX]

APPLID=DBDCKIXX

 

Modifying the UBBCONFIG File

Make the following configurations in the UBBCONFIG file:

•	Configure the ARTSTRN servers for each CICS region

•	Configure the DDR routing definition appropriately

Suppose you have defined two regions, KIXR and KIXX, as shown in Listing 4‑47. Following is a configuration example.

Listing 4‑48 Example of Modifying UBBCONFIG

*GROUPS

GRPKIXR

GRPNO=11

TMSNAME="TMS_ORA"

TMSCOUNT=2

OPENINFO="Oracle_XA:Oracle_XA+Acc=P/yfli/yfli+SqlNet=artkix+SesTm=600+LogDir=/LOGS/xa+DbgFl=0x20"

GRPKIXX

GRPNO=12

TMSNAME="TMS_ORA"

TMSCOUNT=2

OPENINFO="Oracle_XA:Oracle_XA+Acc=P/yfli/yfli+SqlNet=artkix+SesTm=600+LogDir=/LOGS/xa+DbgFl=0x20"

 

*SERVERS

ARTSTRN

SRVGRP=GRPKIXR

SRVID=1101

CONV=Y

MIN=1 MAX=1 RQADDR=QKIXSTRR REPLYQ=Y

CLOPT="-- -s KIXR -l SIMPAPP"

 

ARTSTRN

SRVGRP=GRPKIXX

SRVID=1201

CONV=Y

MIN=1 MAX=1 RQADDR=QKIXSTRX REPLYQ=Y

CLOPT="-- -s KIXX -l SIMPAPP"

 

 

*SERVICES

DEFAULT: SVCTIMEOUT=0 TRANTIME=80

SB00 ROUTING=APPLID

SB01 ROUTING=APPLID

SB02 ROUTING=APPLID

SB03 ROUTING=APPLID

 

*ROUTING

APPLID FIELD=CX_APPLID RANGES="'DBDCKIXX':GRPKIXX,*:GRPKIXR" BUFTYPE="FML32"

 

In this example, requests from KIXX region are routed to ARTSTRN in GRPKIXX, and all other requests are routed to ARTSTRN in GRPKIXR.

Implementing Submitting JCL Online

On z/OS, CICS programs can submit JCL by the WRITEQ TD command and pass the JCL statements to JES internal reader by TDQ. ART CICS Runtime supports this function by using the special TDQ definition and the internal service advertised by TuxJES system.

Before using this feature, make sure ART Batch Runtime and TuxJES environment is set up. For more information, refer to Using Tuxedo Job Enqueueing Service (TuxJES).

Configuring the UBBCONFIG File

The submitted JCL statements are transferred to TuxJES by the ARTTDQ server. To activate this server, configure ARTTDQ in the *SERVERS section in the UBBCONFIG file. Following is an example.

Listing 4‑49 Example of Configuring ARTTDQ in UBBCONFIG

*SERVERS

…

ARTTDQ

SRVGRP=GRP02

SRVID=50

MIN=1 MAX=1

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_strn -e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -s KIXR -L LIST1"

 

Configuring tdqextra.desc

To implement the submitting JCL function, you need to specify the following fields in the tdqextra.desc configuration file:

•	BLOCKFORMAT: An unblocked or blocked record format for the extrapartition queues that are used as the interface to the TuxJES internal reader.

•	INTRDR: Set the TDQ definition as the internal reader.

For example:

IRDR;SIMPAPP;ON LINE SUBMIT JOB;EXTRAQJ; ; ; ;V; ;32767;OUTPUT;DSN; ; ;Y;U;

Where:

Y

Indicates this is an internal reader TDQ.

U

Indicates the block format is UNBLOCKED.

For more information, refer to TD Queue Extra Partition Configuration File in Oracle Tuxedo Application Runtime for CICS Reference Guide.

Implementing Printing CICS Runtime Applications Data

ART CICS Runtime supports printing applications data to a CICS 3270 printer using two methods:

•	Printing with a START command

•	Printing with transient data

General Configurations

Before using either of the methods to implement printing, you need to perform the following configuration tasks:

1.	Configure the printer terminal definition in typeterms.desc. Following is an example:

Listing 4‑50 Example of Configuring Printer Terminal Definition in typeterms.desc

[typeterm]

name=IBM-3287-1

color=YES

defscreencolumn=80

defscreenrow=24

description="IBM 327x family printer"

hilight=YES

logonmsg=NO

outline=NO

swastatus=ENABLED

uctran=NO

userarealen=100

 

Note:

ART CICS does not support the alternate size for printer, so the following attributes must not be defined in typeterms.desc for IBM-3287-1: scrnsize=alternate, altscreenrow, and altscreencolumn.

2.	Configure the printer transactions to the class that will never be executed concurrently:

For example:

TRCLASS1;UNIGRP; A tranclass bidon for UNIGRP; 1

3.	Configure the printer program definition to the ARTSTR1 program group in transactions.desc.

For example:

PRNT;UNIGRP;pg for ARTSTR1; PRNTPROG; ; ; ; ; ; ;ENABLED; ; ; ;TRCLASS1

4.	Configure the printer program to the ARTSTR1 program group in programs.desc.

For example:

PRNTPROG;UNIGRP;pg for UNIGRP;COBOL; ;ENABLED

5.	Define a printer terminal in terminals.desc to indicate the printer LUNAME and TERMID.

For example:

[terminal]

name=PRT1

netname=CICSPRT1

group=UNIGRP

6.	Configure the printer program group to the ARTSTR1 program group using CLOPT -l option in UBBCONFIG.

For example:

CLOPT="-o stdout_str1 -e stderr_str1 -r -- -s KIXR -l UNIGRP"

7.	Do one of the following to specify a printer device:

•	If you use a wc3270 client as the printer terminal, configure the terminal type to IBM-3287-1.

For example:

-tn IBM-3287-1

•	If you use a PCOM client as the printer terminal, configure a LUNAME in the Telnet3270 interface and set the Session Type to Printer in the Session Parameters interface, as shown in following figures:

Figure 4‑7 Configure a LUNAME

Figure 4‑8 Setting the PCOM Session Type

Implementing Printing with a START Command

Figure 4‑9 Printing with a START Command

Figure 4‑9 shows a typical user case. The procedure to implement printing with a START command in such scenario are as follows:

1.	Configure the Pcomm terminal to connect to an external printer device.

2.	Establish the connection from the Printer terminal to ART TCP with the LUNAME you specified previously.

3.	Establish the connection from the Display terminal to ART CICS, and start a transaction A from the Display terminal.

4.	Transaction A invokes an asynchronous transaction B with the START command, and specifies the TERMID as one of the LUNAME defined previously.

5.	Transaction B invokes a BMS or terminal control command, then sends the command application data to the printer terminal, and finally the Printer terminal sends the print job to the connected external printer device for printing.

Implementing Printing with Transient Data

To implement printing with transient data, do the following:

1.	Configure ATIFACILITY and FACILITYID in tdqintra.desc, as follows:

Listing 4‑51 Example of Configuring ATIFACILITY and FACILITYID in tdqintra.desc

# TDQUEUE;GROUP;DESCRIPTION;RECOVSTATUS;TRANSID;TRIGGERLEVEL;USERID;WAIT;WAITACTION;QSPACENAME;TRT;ATIFACILITY;FACILITYID

MTI1;SIMPAPP;TDQ FOR PRINTER;;BBBB;1;;;;;;T;PRT2

 

2.	Define PRT2 in terminals.desc.

For example:

[terminal]

name=PRT2

netname=CICSPRT2

group=UNIGRP

3.	After ART CICS is started, establish the connection between PCOMM (printer with the specified LUNAME= CICSPRT2) and ART CICS.

4.	For ATI users, configure a /Q as follows:

qcreate MTI1 fifo none 2 30 1m 0m "TDI_TRIGGER -q queue_name -d space_name"

When the ATI trigger level is reached, TDI_TRIGGER client will be invoked, and the -q and -d options will notify ART CICS to start a transaction associated queue_name and space_name on the terminal PRT2.

Implementing Invoking Web Services from CICS Applications

ART CICS provides support for invoking web services from CICS applications using the INVOKE WEBSERVICE command. To implement this feature, you need to perform the configuration tasks described in the following sections.

Converting WSDL File into MIF

Convert your WSDL file into the Tuxedo metadata repository input file (MIF) using the Oracle SALT command utility, wsdlcvt. For more information, refer to Configuring an Oracle SALT Application.

Generating VIEW32 Definition from MIF

ART CICS provides a utility, Mif2View32, to generate VIEW32 definition according to MIF and updates the given FML32 definition file. The generated VIEW32 definition keeps the same structure as MIF.

Run the following command:

Mif2View32 -i miffile -o viewfile -f fml32file -w webservice.desc [-v]

Where:

-i miffile

Specifies the metadata input file name.

-o viewfile

Specifies the output VIEW32 definition file name.

-f fml32file

Specifies the updated FML32 definition file.

-w webservice.desc

Specifies the webservice.desc configuration file.

-v

Indicates whether to output details or not.

For more information, refer to Mif2View32(1).

Generating Copybook from VIEW32 Definition (Optional)

For applications that are created in ART CICS, run viewc32 –C –S viewfile to generate the COBOL copybook from the generated VIEW32 definition. For more information, refer to viewc, viewc32(1) in Oracle Tuxedo Command Reference.

Modifying VIEW32 Definition Manually (Optional)

For certain types of applications that are migrated from mainframe, you need to modify the VIEW32 definition fields and adjust the offset manually to enable one-to-one mapping between your copybook and VIEW32 definition.

Following instructions apply to the COBOL copybook that has relevant characteristics.

Check the following and make the modifications accordingly:

•	Container fields in copybook

In copybook, if there are container fields ended with -cont, for example:

09 privateData-cont PIC X(16).

The corresponding item in VIEW32 definition might look like:

struct privateData_v privateData 100 - - 0

To map to your copybook, change that line to:

string privateData1 privateDataList_cont 1 - 16 '\0'

•	Nillable attributes in copybook

In the WSDL schema definition, if there are some items containing the attribute nillable="false", for example:

<element name="walletId" nillable="false" type="xsd:string"/>

In copybook, there should be an extra item describing the attribute:

09 attr-nil-walletId-value PIC X DISPLAY.

In VIEW32 definition, since there is no corresponding item, you need to set the offset value to let the next item point to the correct address. For example:

string walletId_v walletId 1 - 255 "-4"

“-4” indicates the VIEW32 structure has four bytes offset from the next item.

•	Alignment rule for copybook

Copybook has different alignment rule with C structure. When the alignment rule causes an offset between the copybook and VIEW32 definition, you need to set the offset value in VIEW32 definition appropriately.

•	Extra items in VIEW32 definition

In VIEW32 definition, there are helper items indicating the occurrences number. For example:

short payment_v_times payment_times 1 - - 0

In copybook, since there is no corresponding item, remove such items or set the offset value in VIEW32 definition to let the next item point to the correct address.

Here is an example. Suppose you have a copybook like Listing 4‑52, and you have generated a VIEW32 definition as shown in Listing 4‑53, to map the VIEW32 definition to copybook, you need to modify the VIEW32 to Listing 4‑54.

Listing 4‑52 Copybook Example

03 doAuthorizationRequest.

06 buyer.

...

09 email-length PIC S9(4) COMP-5 SYNC.

09 email PIC X(255).

09 attr-nil-email-value PIC X DISPLAY.

09 shippingAdress.

12 attr-nil-shippingAdres-value PIC X DISPLAY.

12 name-length PIC S9(4) COMP-5 SYNC.

12 name PIC X(255).

...

09 walletId-length PIC S9(4) COMP-5 SYNC.

09 walletId PIC X(255).

09 attr-nil-walletId-value PIC X DISPLAY.

06 privateDataList.

09 attr-nil-privateDataLi-value PIC X DISPLAY.

09 privateData-num PIC S9(9) COMP-5 SYNC.

09 privateData-cont PIC X(16).

 

Listing 4‑53 VIEW32 (Before) Example

VIEW buyer_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

...

short email_v_times email_times 1 - - 0

string email_v email 1 - 255 '\0'

short shippingAdress_v_times shippingAdress_times 1 - - 0

struct shippingAdress_v shippingAdress 1 - - 0

...

short walletId_v_times walletId_times 1 - - 0

string walletId_v walletId 1 - 255 '\0'

END

 

VIEW shippingAdress_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

short name_v_times name_times 1 - - 0

string name_v name 1 - 255 '\0'

...

END

 

VIEW privateData_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

short key_v_times key_times 1 - - 0

string key_v key 1 - 255 '\0'

short value_v_times value_times 1 - - 0

string value_v value 1 - 255 '\0'

END

 

VIEW privateDataList_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

short privateData_v_times privateData_times 1 - - 0

struct privateData_v privateData 100 - - 0

END

 

VIEW doAuthorizationRequest_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

short buyer_v_times buyer_times 1 - - 0

struct buyer_v buyer 1 - - 0

short privateDataList_v_times privateDataList_times 1 - - 0

struct privateDataList_v privateDataList 1 - - 0

END

 

Listing 4‑54 VIEW32 (After) Example

VIEW buyer_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

...

short email_v_times email_times 1 - - 0

string email_v email 1 - 255 "-2"

struct shippingAdress_v shippingAdress 1 - - 0

...

short walletId_v_times walletId_times 1 - - 0

string walletId_v walletId 1 - 255 "-4"

END

 

VIEW shippingAdress_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

short name_v_times name_times 1 - - 0

string name_v name 1 - 255 '\0'

...

END

 

VIEW doAuthorizationRequest_v

#TYPE CNAME FBNAME COUNT FLAG SIZE NULL

struct buyer_v buyer 1 - - 0

short privateData_v_times privateData_times 1 - - 0

string privateData1 privateDataList_cont 1 - 16 '\0'

END

 

Configuring webservice.desc

Specify NEWCPY and TRANSACTION in the webservice.desc configuration file. Following is an example.

Listing 4‑55 Example of Specifying NEWCPY and TRANSACTION in webservice.desc

[DFH0XCMNOperation]

REQUEST=DFH0XCMNOperation_v

RESPONSE=DFH0XCMNOperationResponse_v

NEWCPY=Y

TRANSACTION=N

 

Modifying UBBCONFIG

Configure the TMMETADATA and GWWS servers in the UBBCONFIG file. Following is an example.

Listing 4‑56 Example of Adding TMMETADATA and GWWS in UBBCONFIG

*SERVERS

...

TMMETADATA SRVGRP=GROUP2 SRVID=2 CLOPT="-A -- -f pmu.repos"

GWWS SRVGRP=GROUP2 SRVID=3 CLOPT="-A -r -- -iGWWS1"

 

CICS Runtime Logs

Tuxedo System Log

Like other Tuxedo applications, CICS Runtime is managed by Tuxedo that records certain events and problems in a dedicated system log.

This log is the standard Tuxedo User Log (ULOG) whose name is contained in the system variable ULOGPFX of the Tuxedo ubbconfig file.

Example:

ULOGPFX="/home2/work9/demo/Logs/TUX/log/ULOG"

The CICS Runtime Server Logs

When declaring a service in the Tuxedo ubbconfig file, each server has CLOPT options defined including two files:

•	-o option for stdout (normal messages)

The name of this file is stdout_<server name> without the ART prefix.

For example: the ARTSTRN server has a standard output named stdout_strn.

•	-e option for stderr (error messages)

The name of this file is stderr_<server name> without the ART prefix.

For example: the ARTSTRN server has an error output named stderr_strn.

The different stdout and stderr message files for each CICS Runtime server are:

 

Table 4‑7 Message Files by Server

Server name	-o standard output file	-e standard error file
ARTTCPL	stdout_tcp	stderr_tcp
ARTCNX	stdout_cnx	stderr_cnx
ARTSTRN	stdout_strn	stderr_strn
ARTSTR1	stdout_str1	stderr_str1
ARTATRN	stdout_atrn	stderr_atrn
ARTATR1	stdout_atr1	stderr_atr1
ARTTSQ	stdout_tsq	stderr_tsq
ARTDPL	stdout_dpl	stderr_dpl

Note:

In the stderr file of a server all the configuration files mounted are described. The stderr file contains not only the error messages concerning problems encountered when the servers are booted but also information about the different resources loaded. Specifically you will find:

•	The groups of resources installed depending on the -l list parameter of each CICS Runtime server.

•	The resources successfully installed and available for use (remember that an installed resource may be disabled for use) depending on the valorization of each .desc configuration file.

Listing 4‑57 Example of the stdout_strn Just After Start Up for a ARTSTRN Server

Groups loaded: <0001>

|----------|

| GROUP |

|----------|

|SIMPAPP |

|----------|

ARTSTRN: Read config done

|---------------------------------------------------|

| TRANCLASS loaded : < 2> |

|---------------------------------------------------|

| TRANCLASS | GROUP |MAXACTIVE|

|------------------------------|----------|---------|

|TRCLASS1 |SIMPAPP | 001|

|TRCLASS2 |SIMPAPP | 002|

|---------------------------------------------------|

|--------------------------------------------------|

| PROGRAMS loaded : < 4> |

|------------------------------------------------------------------|

| PROGRAM | GROUP |LANGUAGE|EXEC| STATUS |

| | | |KEY | |

|------------------------------|----------|--------|----|----------|

|RSSAT000 |SIMPAPP |COBOL |USER|ENABLED |

|RSSAT001 |SIMPAPP |COBOL |USER|ENABLED |

|RSSAT002 |SIMPAPP |COBOL |USER|ENABLED |

|RSSAT003 |SIMPAPP |COBOL |USER|ENABLED |

|------------------------------------------------------------------|

|---------------------------------|

| TRANSACTIONS loaded : < 4> |

|----------------------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

| | | | |C|C| |R|R| | |T| | | |

|TRAN| GROUP | PROGRAM |ALIA|M|O|PRI|E|E| STATUS |TASK |R| TRAN | TWA |MAX|

| | | | |D|N| |S|S| |DATA |A| CLASS | SIZ |ACT|

| | | | |S|F| |S|T| |KEY |C| | |IVE|

|----|----------|------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

|SA00|SIMPAPP |RSSAT000 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA01|SIMPAPP |RSSAT001 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA02|SIMPAPP |RSSAT002 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA03|SIMPAPP |RSSAT003 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|-----------------------------------------------------------------------------------------------------|

Warning: zero TSQMODEL loaded!!

FILES<FILE> lineNo(1) skipping Record: Group not to load

FILES<FIC3> lineNo(4) skipping Record: Group not to load

 

We can note in this example that

•	One group (SIMPAPP) is selected with the -l option

•	Four configurations files are used: transactions, tranclasses, programs and tsqmodels.

•	Information on the successful loading of these resources (Warning: zero TSQMODEL loaded).

•	The detail of the resources loaded and their explicit characteristics (name, group, description …) even default/implicit values were used in the .desc file leaving the fields filed with space(s).

Disabling and Enabling Programs

Sometimes, problems are encountered in a program that significantly impacts your system and the program must be eliminated urgently by prohibiting end-users from running it. In the immediate, this helps temporarily to stabilize the system giving time to analyze and solve the dysfunction.

As on z/OS, CICS Runtime allows to disable a program. A program is disabled by modifying the CICS Runtime configuration file programs.desc. This file contains a dedicated field, the STATUS field, to indicate if a program is DISABLED or ENABLED (status by default).

See also dynamic administration of CICS resources information in the Oracle Tuxedo Application Runtime for CICS Reference Guide.

Disabling Programs

To switch your transaction from enabled to disabled, you have to modify the seventh field of this csv file, to change the previous value from an implicit (" " space(s)) or an explicit ENABLED status to the explicit DISABLED status.

After shutting down and booting the CICS Runtime Tuxedo servers, your modifications of one or more programs will be taken in account.

If you disable a program, when somebody wants to use it, the error messages displayed depend on the way that the application handles CICS errors.

Listing 4‑58 Example Simple Application SA02 COBOL Program Set to DISABLED in programs.desc

#PROGRAM;GROUP;DESCRIPTION;LANGUAGE; ; ;STATUS

RSSAT000;SIMPAPP; Home Menu Program of the Simple Application ;COBOL

RSSAT001;SIMPAPP; Customer Detailed Information Program of the Simple Application ;COBOL; ; ;ENABLED;

RSSAT002;SIMPAPP; Customer Maintenance Program of the Simple Application;COBOL; ; ;DISABLED;

RSSAT003;SIMPAPP; Customer List of the Simple Application ;COBOL

 

Enabling Programs

To enable a program, you have only to do the opposite, changing the STATUS field from DISABLED to ENABLED or " " (at least one space).

After shutting down and booting the CICS Runtime Tuxedo servers, your modifications of one or more programs take effect.

Checking the Change in Program Status

If you consult the logs of the different transactions servers or the CICS Runtime you will note the modification of the modified status in the stderr_* logs.

Just after the start up of this server, the logs shows (in italics) that this program is disabled.

Listing 4‑59 Log Report Showing Program Status

Groups loaded: <0001>

|----------|

| GROUP |

|----------|

|SIMPAPP |

|----------|

ARTSTRN: Read config done

|---------------------------------------------------|

| TRANCLASS loaded : < 2> |

|---------------------------------------------------|

| TRANCLASS | GROUP |MAXACTIVE|

|------------------------------|----------|---------|

|TRCLASS1 |SIMPAPP | 001|

|TRCLASS2 |SIMPAPP | 002|

|---------------------------------------------------|

|--------------------------------------------------|

| PROGRAMS loaded : < 4> |

|------------------------------------------------------------------|

| PROGRAM | GROUP |LANGUAGE|EXEC| STATUS |

| | | |KEY | |

|------------------------------|----------|--------|----|----------|

|RSSAT000 |SIMPAPP |COBOL |USER|ENABLED |

|RSSAT001 |SIMPAPP |COBOL |USER|ENABLED |

|RSSAT002 |SIMPAPP |COBOL |USER|DISABLED |

|RSSAT003 |SIMPAPP |COBOL |USER|ENABLED |

|------------------------------------------------------------------|

|---------------------------------|

| TRANSACTIONS loaded : < 4> |

|----------------------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

| | | | |C|C| |R|R| | |T| | | |

|TRAN| GROUP | PROGRAM |ALIA|M|O|PRI|E|E| STATUS |TASK |R| TRAN | TWA |MAX|

| | | | |D|N| |S|S| |DATA |A| CLASS | SIZ |ACT|

| | | | |S|F| |S|T| |KEY |C| | |IVE|

|----|----------|------------------------------|----|-|-|---|-|-|----------|-----|-|--------|-----|---|

|SA00|SIMPAPP |RSSAT000 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA01|SIMPAPP |RSSAT001 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|SA02|SIMPAPP |RSSAT002 | |N|N|001|N|N| ENABLED |USER |Y| |00000|999|

|SA03|SIMPAPP |RSSAT003 | |N|N|001|N|N|ENABLED |USER |Y| |00000|999|

|-----------------------------------------------------------------------------------------------------|

Warning: zero TSQMODEL loaded!!

 

Removing and Adding Applications for CICS Runtime

Sometimes, you want to delete an application from a given machine either to definitely delete all its components or to move them to another machine. If all the resources used by your application were defined in one or more resource groups dedicated to your application, you have only to suppress these groups from CICS Runtime and eventually install them elsewhere.

Each CICS Runtime Tuxedo Server reads a list of groups to be selected and installed at start up, contained in its CLOPT options after the -l parameter. To remove or add group(s) from an application, you have only to remove or add theses groups from this list for each CICS Runtime Tuxedo server.

Your modifications on one or more programs take effect after shutting down and booting up the CICS Runtime Tuxedo servers.

Listing 4‑60 Example of Application in ARTSTRN Server

ARTSTRN SRVGRP=GRP02

SRVID=20

CONV=Y

MIN=1 MAX=1 RQADDR=QKIX110 REPLYQ=Y

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_strn

-e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -

s KIXR -l SIMPAPP"

 

If you want to add one or more groups, you have to concatenate these new groups to those previously defined, separating them with a ":" character.

Listing 4‑61 Example of Adding group1 and group2 in ARTSTRN Server

ARTSTRN SRVGRP=GRP02

SRVID=20

CONV=Y

MIN=1 MAX=1 RQADDR=QKIX110 REPLYQ=Y

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_strn

-e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -

s KIXR -l SIMPAPP:GROUP1:GROUP2"

 

If you want to remove groups, you remove them from the -l lists when they are present, leaving only one : character between the remaining groups.

Listing 4‑62 Example of Removing group1 in ARTSTRN Server

ARTSTRN SRVGRP=GRP02

SRVID=20

CONV=Y

MIN=1 MAX=1 RQADDR=QKIX110 REPLYQ=Y

CLOPT="-o /home2/work9/demo/Logs/TUX/sysout/stdout_strn

-e /home2/work9/demo/Logs/TUX/sysout/stderr_strn -r -- -

s KIXR -l SIMPAPP:GROUP2"

 

Notes:

•	When the groups are removed, all the resources of these groups are only logically suppressed. If you want also to suppress them physically, you have to delete all the lines of the CICS Runtime resource configuration files containing the group names.

•	When the groups are added, all the resources of theses groups must be present in the different CICS Runtime resource configuration files under the group names. To avoid future problems, do not omit to declare resources in a group because they are already declared in groups from other applications.

•	When groups are added or removed, be careful to indicate the same list of groups in each CICS Runtime server.

CICS Runtime C Program Support

ART CICS allows users to implement and run CICS applications in C language.

Running C Program in CICS Runtime

Each C program is loaded as a COBOL program and executed in COBOL runtime, so CICS C program support is COBOL production depended.

C Programming Restrictions and Requirements

Restrictions and requirements for CICS/C support on ART CICS are listed as below.

•	The EXEC CICS commands related to nonstructured exception handling are not supported, including:

•	HANDLE ABEND

•	HANDLE AID

•	HANDLE CONDITION

•	IGNORE CONDITION

•	PUSH HANDLE

•	POP HANDLE

•	In a C application, every EXEC CICS command is treated as if it had NOHANDLE or RESP option specified.

•	On Linux, file names and words are case-sensitive. If a header file name is in lowercase in C source file while such name in file system is in uppercase, errors will occur in compiling.

•	Does not support packed decimal data.

•	Does not support the use of CICS command in macros.

•	Support CICS keywords in mixed case.

•	Where CICS expects a fixed-length character string (such as a program name, map name, or queue name), you must pad the literal with blanks up to the required length if it is shorter than expected.

•	Take care not to define a variable with a field name. That behavior will cause C compiler abend.

•	/**/ is used for single line comments. Do not put a comment in the middle of an EXEC CICS command.

•	ART CICS does not support argc, argv, and envp.

•	ART CICS provides two methods to access COMMAREA:

•	ADDRESS COMMAREA

•	Provide an extern global pointer __commptr declared by ART CICS pre-processor automatically. __commptr is system reserved; users cannot define it as other usages.

•	ART CICS provides two methods to access EIB:

•	ADDRESS EIB

•	Provide an extern global pointer __eibptr declared by ART CICS pre-processor automatically. __eibptr is system reserved; users cannot define it as other usage.

•

The EIB declarations are enclosed in #ifndef and #endif lines, but are not included in all translated files. ART CICS publishes header file dfheiblk.h to contain the definition of all the fields in the EIB. Each translated file just needs to include this header file and all actions are completed by pre-processor automatically.

•	BMS screen attributes definitions: C versions of the DFHBMSCA and DFHAID files are supplied by CICS, and may be included by the application programmer when using BMS.

•	The string handling functions in the C standard library use a null character as an end-of-string marker. Because CICS does not recognize a null as an end-of-string marker, customers must take care of it when using C functions, for example strcmp, to operate on CICS data areas.

•	ART CICS provides iscics() declared in cics.h as well, but users only need to modify makefile to include the header file path.

•	Keep EXEC CICS as a whole in one line.

•	Multiple CICS commands in one line is not support.

•	#pragma will be automatically translated to comments.

•	C function exit() will be translated to return.

•	Keep C function main(), its parameter list, and parenthesis in one line. For example,

void main(int argc, char **argv)

•	COMMAREA should be a pointer. ART CICS only supports specifying a struct by pointer, not value.

Accessing EIB from C

The address of the EXEC interface block (EIB) is not passed as an argument to a C main function; however, users can use the following two methods to obtain the address of the EIB:

•	Using ADDRESS EIB

•	Using global pointer __eibptr which points to EIB

Accessing COMMAREA from C

The address of COMMAREA is not passed as an argument to a C main function; however, users can use the following two methods to obtain the address of the COMMAREA:

•	Using ADDRESS COMMAREA

•	Using global pointer __commptr which points to COMMAREA

CICS Command Translator

ART CICS provides prepro-cics-C.pl for CICS/COBOL APIs translation. For more information about prepro-cics-C.pl, please refer to prepro-cics-C.pl.

C Program Compilation

In order to make sure the C programs could be successfully loaded by COBOL runtime, please build C programs using COBOL compiler rather than gcc/g++.

Use cob to compile C source code as a callable shared object. Please note that dynamic library must have the same name as the C source file name in uppercase.

For example,

CPYINC=../includes

cob -z,CC zample_treated.c -o ZAMPLE.so -CC -I${CPYINC}