CICS related content
If any form of CICS journaling or logging is required and any form of CICS start other than INITIAL then the machine that CICS TS is running in MUST be in some form of SYSPLEX configuration if only a MONOPLEX. CICS now uses the zOS logger and has done away with its own journaling (DFHJCT tables and DFHRSD datasets etc) although it can be configured to use DASD-only logs rather than logging to a Coupling Facility.
CICS System Log (DFHLOG)
This is used for internal CICS logging and is the rough equivalent of the old DFHJ01 plus some more.
Used for recovery purposes (DTB, emergency restart).
CICS needs it to backout recoverable resources, COLD starts (recover a conversation state), WARM starts (pre-shutdown state), EMERGENCY start….
CICS connects to the system log automatically during initialisation.
Default names: region_userid.applid.DFHLOG and ….DFHSHUNT
Whenever a new set of System Logs are defined, or if they are to be cleared down CICS must be started with START=INITIAL.
Example to define the CICS System log streams….
//SYS013 EXEC PGM=IXCMIAPU
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DATA TYPE(LOGR) REPORT(YES)
DELETE LOGSTREAM NAME(CICS.DFHLOG.MODEL)
DEFINE LOGSTREAM NAME(CICS.DFHLOG.MODEL)
MODEL(YES)
DASDONLY(YES)
LOWOFFLOAD(40)
HIGHOFFLOAD(80)
STG_SIZE(9094)
MAXBUFSIZE(64000)
DELETE LOGSTREAM NAME(CICS.DFHSHUNT.MODEL)
DEFINE LOGSTREAM NAME(CICS.DFHSHUNT.MODEL)
MODEL(YES)
DASDONLY(YES)
MAXBUFSIZE(64000)
/*
DFHLGLOG is the CICS log of logs and is used for Forward Recovery products like CICSFR. We, and most other sites, have defined it as DUMMY.
Unless otherwise specified a HLQ of IXGLOGR is prefixed to the logstream name defined in CICS.
May need to define an alias for IXGLOGR and make definitions in TSS.
Check SMS definitions.
Some thought should be given to adequate sizing for a production workload.
CICS supplies a utility (DFHLSCU) to read old style production logs and to predict efficient parameters for your new logstreams.
Deleting and listing of logstreams should also be done using the IXCMIAPU utility.
User Journals and Forward Recovery logs
These are defined in a similar way. They can / should be defined to manage themselves (e.g. keep data for a specific time). Due to the nature of the processing at Dixons it is unlikely that any such logs will need to be defined.
Abstract Availability for CICS® Transaction Server for z/OS® V3.2 is planned for 29 June 2007. CICS TS V3.2 delivers significant innovation for application connectivity, application reuse, and service management. The major new functions in CICS TS V3.2 fall into three main areas: Application connectivity New Web services capabilities include support for recent standards, interoperability profiles, and the ability to send large amounts of binary data efficiently. The ability to interconnect CICS regions over TCP/IP for Distributed Program Link (DPL) allows exploitation of System z™ IP networking, and provides an alternative to Systems Network Architecture (SNA). All CICS TCP/IP workloads benefit from simple and robust systems and workload management facilities. Application reuse Enhancements to the efficiency of deploying Web services into CICS include handling a wider range of programming language and XML data types and attributes, and improvements to performance and diagnostics. Applications will now automatically utilize 64-bit storage for data held in containers, providing for improved scalability and performance. The CICS integrated translator is now supported by the z/OS XL C/C++ compiler, delivering more comprehensive development and debug capabilities. Service management Online management of program libraries makes it easier to maintain continuous system availability while bringing new or changed applications into production. Support for Enterprise Workload Manager (EWLM) makes possible end-to-end workload monitoring in heterogeneous environments. Potential for performance enhancements is offered by making threadsafe the core APIs for accessing local and RLS VSAM files, journals, WebSphere® MQ, and system autoinstalled Global User Exits. There is improved management of Java™ workloads. The CICSPlex® SM Web User Interface has new help and map capabilities and usability enhancements. Larger amounts of data can now be stored in shared data tables (>2 GB) and VSAM entry sequenced data set (ESDS) files (>4 GB). Improvements to monitoring and statistics facilities include more precise timing data and compression of monitoring records. The default internal trace table size has been increased for better problem diagnosis. T he limit on the number of CICS regions supported in a sysplex is increased by enabling a CICS region to join a named XCF group.
“not a lotta people know that…” CICS comes with a valuable little friend FOR FREE.
When you installed CICS you also installed (or had the opportunity to install) a VERY USEFUL tool which has a misleading name – CICSPlex System Manager (CPSM)
CPSM is often overlooked for two reasons…
1. CICSPlex is misleading; it relates to ANY collection of CICS regions not necessarily in a SYSPlex
2. Setting up and using CPSM is not well documented. If you are running just a few or many CICS regions and can afford to run 1 or 2 more then there may be a lot of benefits including possible software savings in setting up and using CPSM.
Benefits
From a single Web Browser or ISPF session you can
• Display or change the characteristics of any / all CICS region(s)
• Define and propagate CICS resources
• Define and perform Real Time Analysis (RTA)
• Define and perform monitoring (MON)
• Define Workload Management specifications (WLM) allowing the system to dynamically decide where a transaction will run.
• You may already be running ISV software to do some of these things that come at no extra cost to your Company. (It has to be said that there is no extra data collection involved so CPSM does NOT replace any reporting you might do from other CICS monitoring products. If you can live without historic reporting then direct replacement might be a possibility.)
Functions of CPSM
A single-system image (SSI)
CPSM can provide a real-time, single-system image (SSI) of all CICS regions and resources that make up a Company’s entire transaction processing environment. CPSM creates a topology of the CICS systems and resources mapping the relationships between them. A technician no longer needs to know the location of a resource before working with it. The SSI applies to all CPSM applications: operations, monitoring, real-time analysis (RTA), and workload management (WLM).
A single point of control.
A CPSM user can manage all participating CICS systems and their resources from a single session, a single point of control for the enterprise so the CPSM operator is able to manage large and complex configurations of CICS resources from a single session.
Management of your business applications
Your entire enterprise can be managed in terms of your business applications rather than your CICS systems. Using Business Application Services (BAS) you can manage your resources at the application level, by providing:
• A single point of definition for resources. All the resources for the CICS systems and the relationships between them are held on the CPSM data repository. CPSM produces local and remote instances of a resource from a single definition.
• Logical scoping, which enables you to link and manage your resources according to their business relationship, rather than by their physical location in a CICS system.
• Installation of resources, either automatically at CICS start up, or dynamically, as required, into the participating CICS regions. BAS provides an alternative to resource definition online (RDO). BAS frees resource definitions from associations with only one group. Resource definitions can be used and reused, and associated with more than one group as needed. You can, if you wish, manage individual resources, rather than a group.
Operations for the entire CICSplex
Due to SSI a user can (from any point of control) perform tasks either across the entire CICSplex, across a selected subset or for an individual CICS region.
Information can be displayed about the current status of one or more instances of a CICS resource in an enterprise.
The status of resources can also be changed with a single entry.
Action commands can be issued to affect the status of the displayed resources.
You use the Web User Interface or the ISPF EUI to issue the commands. The responses are displayed in panels known as operations views that can summarize related facts and provide access, using hyperlinks to other, related information.
Management of your workloads
CPSM's workload management (WLM) uses a supplied dynamic routing program (EYU9XLOP) to route eligible work requests from a requesting region to a suitable target region selected at the time the transaction is initiated.
EYU9XLOP can be customised and used in conjunction with the CICS-supplied user-replaceable programs DFHDYP and DFHDSRP to handle most dynamic routing requirements.
Automated exception reporting for CICS resources
CPSM's Real-Time Analysis (RTA) function provides automatic, external, notification of any nominated conditions. The notification can take the form of a console message, or of an alert to NetView, or both.
Real-time analysis can asked to notify about any aspect of a resource's status.
Real-time analysis also enables you to take actions without having to use an external automation product.
Monitoring functions for the collection of statistical data for CICS resources
The CPSM monitoring functions support the collection of performance-related data, at user-defined intervals, for named resource instances within a set of CICS systems.
An application programming interface (API) CPSM provides an application programming interface (API) that allows applications to:
• Access information about CICS and CPSM resources
• Invoke the services of CPSM. A command-level interface is available to programs that are written in Assembler, PL/I, COBOL or C. In addition, a REXX run-time interface is available under z/OS Batch, TSO, and NetView.
Management of the CPSM environment
You manage the CPSM environment using:
• CPSM objects To define the configuration of your CICS systems to CPSM (and to define your BAS, WLM, RTA, and monitoring requirements) you create CPSM objects, and associate them with each other. For each object, and for each association or link between them, a record is created in a CPSM data repository.
• Data repository The data repository contains the objects that define the CPSM components, resources, system management requirements, and the relationships between them. The definitions can be created using the Web User Interface or EUI views, the CPSM API, or the batched repository-update facility.
• The batched repository-update facility The batched repository-update facility allows you to create and update large numbers of CPSM and CICS resource definitions by submitting one command that is used as the template for other definitions. The batched repository-update facility is also used for migrating your definitions from one platform to another, and for backing up the data repository. More information is given in The batched repository-update facility.
Management of time-dependent activity
CPSM's facilities can be made time dependent. RTA and monitoring definitions can be activiated only during a specific time for instance. You are able to create time period-definitions that control:
• Exactly when any part of your enterprise is operational, regardless of the local time zone.
• The times you want certain system management functions to be operational. Integration with Tivoli(R) Business Systems Manager For users of Tivoli Business Systems Manager, the CPSM Instrumentation feature supplied with that product enables users to view status information for the CMAS and MAS environments of the CICSplex.
The WUI will run fine within an existing CICS region. It is "just an application".
Some things to consider (like with all monitors):
Suggest keeping the WUI separate since it is a systems management tool and is not managed like a "normal" application region.
This is similar to keeping TOR's "IBM code only" regions (OK, maybe your CICS monitor is in there too).
Bring up the WUI(s) at IPL time along with the other CPSM address spaces.
The WUI(s) should run at a higher service level.
Although CPSM and the WUI are not yet a replacement for a CICS monitor, the CPSM/WUI is a tremendous tool for spotting hung transactions, disabled/released resources, dump activity, etc from a single location.
It is also a tremendous tool for a single point of corrective action.
When customized with your own menus and views, you can create a tremendous cross-plex exception monitor that gives you immediate visibility to error conditions no matter where they come from.
The ability to issue cross plex commands and see the results frees you from the MVS ROUTE command, the potential WTO buffer shortage problems and the coordinated naming conventions required to exploit it.
CPSM and the WUI are not yet the "be all to end all", but it is a great start and IBM appears to be completely behind the product.
While the "starter set" is nice, the real power comes with an effective customization based on your monitoring requirements. It is easy to customize and well worth the minor investment in time to master the tailoring. The basic menu and view tailoring is easy.
It is also possible to create customized URL's that can be imbedded in other web pages and incorporated in online documentation. You could have a project to update Web based Operations procedures to include step by step procedures that actually link to the WUI (in a frame) for the operator action.
Check software and hardware requirements
Research and apply any required maintenance
Be aware of changes to CICS-supplied transactions
Review changes introduced with this release including the following…
Follow ServerPac dialogs
Integrating CICS and OS390
(Some of these tasks will already be in place for existing CICS also some tasks can wait.)
Review messages and automation
Defining the logger environment for CICS journaling
Applying service to CICS Transaction Server for OS/390
Perform IVPs
Convert a test region and check it out
Tasks required for converting a region include