- ABORT
- Two-phase commit consists of the PREPARE
and COMMIT phases. Within the COMMIT phase, there are two possible actions:
COMMIT and ABORT. The ABORT action for data belonging to full function DL/I
databases is backout. There is no backout for data belonging
to DEDBs, because it has not been written to the database before the COMMIT
phase. The effect of an ABORT on DEDBs is also referred to as an undo. Because a CICS® thread may be accessing data belonging to both full
function DL/I databases and DEDBs, we use the term ABORT to refer to both
backout and undo.
- ACB (application control block)
- Created from the
output of DBDGEN and PSBGEN and placed in the ACB library (ACBLIB) for use
during online and DBD region type execution of IMS™.
- active
- In an XRF environment, active describes
the system that is currently supporting processing requests.
- ADS (area data set)
- A copy of a DEDB area. You
can have up to seven copies of the same area, which are all automatically
maintained in synchronization.
- AGN (application group name)
- DBCTL views the set
of PSBs that can be accessed by one particular CICS system or BMP as a single entity, known
as an application group. Application groups, and the
names of the resources within those groups, are placed in tables in DBCTL’s
security matrix data set(s) using the IMS security maintenance utility. An AGN for
a CICS system is specified in the DRA startup table. When a CICS system requests
connection to DBCTL, RACF® checks against its resource class table to ensure
that the AGN being specified authorizes it to connect to that DBCTL subsystem.
- alternate
- In an XRF environment, alternate describes
system that is standing by waiting to take over the workload when the active
system fails or a takeover is initiated.
- alternate TP PCB
- An alternate TP (transaction processing)
PCB defines an alternate destination (a logical terminal or a message program)
and can be used instead of the I/O PCB when it is necessary to direct a response
to a terminal. Alternate TP PCBs appear in PSBs used in a CICS-DBCTL environment,
but are used only in an IMS/VS DC or IMS TM environment. CICS applications using DBCTL cannot successfully
issue requests that specify an alternate TP PCB, an MSDB PCB, or a GSAM PCB.
However, a PSB that contains PCBs of these types can be scheduled successfully
in a CICS-DBCTL environment. Alternate PCBs are included in the PCB address
list returned to a call level application program The existence of alternate
PCBs in the PSB affects the PCB number used in the PCB keyword in an EXEC
DLI application program.
- AMODE (addressing mode)
- Refers to whether program
addresses are 24 or 31 bits.
- AOR (application-owning region)
- A CICS address space
whose primary purpose is to manage application programs. It receives transaction
routed requests from a terminal-owning region (TOR). It may also contain file
related resources in a system that does not have a database-owning region
(DOR). See also DOR (database-owning region) and TOR (terminal-owning region).
- APPLID
- Operand of the CICS system initialization table that specifies
the 1- to 8-character application name of a CICS system. It is the name by which the CICS system is to be known to other systems or regions.
- AVM (MVS™ availability manager)
- Handles communication
between active and alternate IMS XRF systems. See also CAVM (CICS availability manager).
- BMP (batch message processing program)
- BMPs are
application programs that perform batch type processing online and can access
databases controlled by DBCTL. You can run the same program as a BMP or as
a batch program.
- call
- An instruction in COBOL, assembler, or PL/I
that is used by an application program to request DL/I services. It does not
require translation. Contrast with command.
- CAVM (CICS availability manager)
- Handles communication
between active and alternate CICS systems in a CICS system with XRF. See also AVM (MVS availability manager).
- CCTL (coordinator control subsystem)
- This refers
to the transaction management subsystem that communicates with the DRA, which
in turn communicates with DBCTL. In a CICS-DBCTL environment, the CCTL is CICS. The term is used in a number of IMS operator commands that apply to DBCTL,
and in the IMS manuals.
- checkpoint
- For applications, a point at which the program commits that the changes it has made to
the database are consistent and complete, and releases database segments for
use by other programs. You can request checkpoints at appropriate points in
a program to provide places from which you can restart that program if it,
or the system, fails.
For systems, a point
in time from which IMS can start again if a failure makes recovery necessary.
The checkpoint is performed by IMS itself.
- CI (control interval)
- The unit of information transmitted
to or from auxiliary storage by VSAM, independent of logical record size.
- CICS monitoring facility
- The CICS monitoring
facility gives a comprehensive set of operational data for CICS, using a data
recording program. Data is normally output to the system in SMF data sets.
- cold start
- The standard initialization sequence
that is performed by the system initialization program without regard for
prior system activity.
- command
- In CICS, an instruction similar in format to
a high-level programming language statement. CICS commands usually include the verb EXECUTE
(abbreviated to EXEC), and can be issued by an application program to make
use of CICS facilities. With DL/I, the format of the command is EXEC DLI. Commands
require processing by the CICS translator. Contrast with call.
- CRC (command recognition character)
- A character
that denotes a DBCTL operator command. DBCTL operator commands have / as their
default CRC. You can override the default CRC on the DBCTL job, but remember
that each DBCTL subsystem within an MVS image must have
a unique CRC and that CRC must be unique with respect to every other
subsystem on the processor, not just DBCTL subsystems. If you are using the
CICS-supplied operator transaction, CDBM, to issue operator commands to DBCTL,
you must use the default CRC, even though your DBCTL is using some other CRC.
- CRLP (card reader/line printer)
- Or in-stream sequential
terminal. Can be used as a means of automating connection to a different DBCTL
or to connect automatically when CICS was not connected to DBCTL at shutdown.
- data availability
- Data availability is an IMS enhancement available
with DBCTL. It allows PSB scheduling to complete successfully even if some
of the full function databases it requires are not available.
- database integrity
- The protection of data items
in a database while they are available to any application program. Protection
includes isolating the effects of concurrent updates to a database by two
or more application programs.
- database organization
- The physical arrangement
of related data on a storage device. DL/I database organizations are hierarchical.
- database record
- (1) A collection of DL/I data elements
known as segments that are hierarchically related to
a single root segment. (2) In a DL/I or IMS database, a root segment and all its descendant
segments.
- database reorganization
- The process of unloading
and reloading a database to optimize physical segment adjacency, or to modify
the DBD.
- data sharing
- An IMS term. Data sharing can be done at:
- Block level, which allows multiple subsystem access
to the same database, controlled by means of a lock manager.
- Database level, which allows application programs
in one IMS subsystem to read data while another program in another IMS subsystem reads
from, or updates, the same database.
- DBCTL
- DBCTL is an interface between CICS Transaction Server for z/OS® and IMS that
allows access to IMS DL/I full function databases and to Data Entry Databases
(DEDBs) from one or more CICS systems without the need for data sharing. It also
provides release independence, virtual storage constraint relief, operational
flexibility, and failure isolation.
- DBD (database description)
- In IMS, the collection
of macro parameter statements that describes an IMS database. These statements describe the
hierarchical structure, IMS organization, segment length, sequence fields, and
alternate search fields. They are assembled to produce database description
blocks.
- DB PCB (database PCB)
- A PCB that supports communication
between an application program and a database.
- DBRC (database recovery control)
- An IMS facility that
maintains information needed for database recovery, generates recovery control
statements, verifies recovery input, maintains a separate change log for database
data sets, and supports sharing of IMS DL/I database by multiple IMS systems.
- DDIR (database directory)
- A list of data management
blocks (DMBs) that define for DL/I the physical and logical characteristics
of databases that are used by application programs. A CICS system initialization
parameter of the same name specifies a suffix to identify a DDIR.
- DEDB (data entry database)
- In IMS, a direct-access
database originally provided in the Fast Path feature. DEDBs can be divided
into independent areas, which increases availability
of data. DEDBs provide a high level of availability for, and efficient access
to, large volumes of data. They are hierarchic structures that contain a special
type of segment called a sequential dependent segment (SDEP) that is used
for fast collection of information and is useful, for example, in journaling
and auditing applications. Applications that access DEDBs can also use subset
pointers, which allow more efficient processing of long segment chains. The
database is accessed using Media Manager, which is a component of Data Facility
Product (DFP).
- DFHDBAT
- The interface between the DRA and CICS. DFHDBAT adapts CICS’s calls to the DRA’s interface when accessing DBCTL databases.
See also DRA (database resource adapter).
- DFHDBCON
- DFHDBCON, the DBCTL connection program,
is invoked during connection to DBCTL.
- DFHDLI
- The CICS DL/I router module. Determines whether
a DL/I request should be processed by local DL/I, remote DL/I, or passed to
DBCTL.
- DFH$INDB
- DFH$INDB, the CICS-supplied
sample in-doubt resolution program helps you decide whether to commit or backout
updates that are in-doubt after CICS has disconnected abnormally from DBCTL.
DFH$INDB produces a list of in-doubts, plus the action needed to resolve
each one.
- DIB (DL/I interface block)
- Whenever you issue an
EXEC DLI command, DL/I responds by storing the information in the DIB in your
program. It is inserted automatically into your program by the CICS translator.
See also DIBSTAT.
- DIBSTAT
- The DL/I status code, which is contained
in the DIB. It indicates the success (or otherwise) of your EXEC DLI command.
- DL/I (Data Language/I)
- A high-level interface between
applications and IMS. It is invoked from PL/I, COBOL, or Assembler language,
or (for command-level only) C language. by means of ordinary subroutine calls.
DL/I enables you to define data structures, to relate structures to the application,
and to load and reorganize these structures. It enables applications programs
to retrieve, replace, delete and add segments to databases. See also command.
- DMB (data management block)
- An IMS control block
that resides in main storage and describes and controls a physical database.
It is constructed from information obtained from the application control block
(ACB) library or the database description (DBD) library.
- domain
- A logical grouping of CICS function; for
example, the storage domain or the monitoring domain.
- DOR (database-owning region)
- A CICS address space
whose primary purpose is to manage files and databases. See also AOR (application-owning region) and TOR (terminal-owning
region).
- DRA (database resource adapter)
- The architected
interface that enables DBCTL databases to be accessed from CICS.
- DRA control exit
- Enables the DRA to pass information
from itself and DBCTL independently of CICS. It is invoked whenever the DRA needs
to determine whether to continue processing, as follows:
- The DRA successfully connects to DBCTL
- DBCTL is terminated normally using /CHECKPOINT FREEZE or /CHECKPOINT PURGE
- An attempt to connect to DBCTL fails
- A CICS INIT request is canceled
- The DRA fails.
- DRA startup parameter table
- The DRA startup parameter
table provides the parameters needed to define a DBCTL subsystem.
- equivalent
- In an XRF environment, equivalent describes
DBCTL subsystems that are defined as members of the same RSE. See also RSE (recoverable service element) and RST (recoverable service table).
- full function databases
- Full function databases
are hierarchic databases that can be accessed using DL/I and can be processed
by batch programs and BMPs.
- function shipping
- The process by which CICS accesses resources
on another CICS system. The process is transparent to application programs. See
also remote DL/I.
- generic APPLID
- The name by which the active-alternate
pair of CICS systems is known to other systems or regions.
- gigabyte
- The exact value 1 073 741 824.
- global user exit
- A global user exit is a point
in a CICS module at which CICS can pass control to a program that you have written
(known as an exit program), and then resume control
when your program has finished. When an exit program is enabled for a particular
exit point, the program is called every time the exit point is reached. Global
user exits used with DBCTL are: XDLIPRE and XDLIPOST, XRMIIN and XRMIOUT;
plus XXDFA, XXDFB, and XXDTO, which are used with XRF. See also task-related user exit.
- HSSP (high speed sequential processing)
- HSSP is
useful with applications that do large scale sequential updates to DEDBs.
It can reduce DEDB processing time, enables an image copy to be taken during
a sequential update job, and minimizes the amount of log data written to the IMS log.
See also DEDB (data entry database).
- IMS monitor
- An IMS monitoring tool, which can be run online,
unlike the IMS DB monitor which can be run in batch only. DBCTL enables CICS users who do
not have an IMS DM/TM system to use the IMS monitor.
- in-doubt
- Refers to a piece of work that is pending
during commit processing. If commit processing fails between polling of subsystems
and the decision to execute the commit, recovery processing must resolve the
status of any work that is in-doubt.
- in-flight
- Refers to a piece of work that is being
processed when a system failure occurs.
- I/O PCB
- An input/output PCB that is needed to issue
DBCTL service requests.
- IRLM (internal resource lock manager)
- A global
lock manager that resides in its own address space, and gives the option of
keeping most of its control blocks in local storage instead of in the common
storage area (CSA). You must use the IRLM to maintain data integrity if you
are sharing databases at block level. (For VSAM databases, a block is a control
interval (CI); for any other kind of database, it is a physical block.) You
also need the IRLM to process a set of common databases from multiple IMS (or CICS Transaction Server for z/OS)
subsystems. You may optionally use the IRLM in a database level sharing environment
for improved integrity for read-only subsystems. The IRLM is also the lock
manager used by DATABASE 2 (DB2®).
- JES (job entry subsystem)
- The subsystem used in CICS with XRF to route commands and queries from the alternate to the active
system.
- journal
- A set of one or more data sets to which
records are written during a CICS run:
- By CICS to implement user-defined resource protection (logging to the system
log)
- By CICS to implement user-defined automatic journaling (to any journal,
including the system log)
- Explicitly by JOURNAL command (or macro) from an application program (user
journaling to any journal including the system log).
- KB (kilobyte)
- The abbreviation KB (as in 1KB) represents
the exact value of 1024.
- linkage editor
- A processing program that prepares
the output of language translators for execution. It combines separately produced
object modules, resolves symbolic cross-references among them, and produces
executable code that is ready to be fetched or loaded into virtual storage.
- local DL/I
- DL/I residing in the CICS address space.
Discontinued in CICS Transaction Server..
- LP (logical partition)
- A partition, in a central
processing complex, capable of running its own MVS image. It comprises a set of hardware resources
(processors, storage, channels, and so on, sufficient to allow a system control
program such as MVS to execute.
- MB (megabyte)
- The abbreviation MB (as in 1MB) represents
the exact value of 1 048 576.
- monitoring
- In CICS, data produced on timing and resources
used by a task or a logical unit of work (UOW).
Note that CICS distinguishes between monitoring and statistics, but IMS does not. See
also statistics.
- multi-MVS environment
- An environment that supports
more than one MVS image. See also MVS image.
- MVS image
- Can be a physical processing system (such as an IBM® 3090), which can be partitioned. Each partition, which has one or more processors,
is an MVS image.
- NODHABEND
- A keyword on the EXEC DLI SCHD PSB command.
Used to prevent DHxx abends being issued after PSB schedule request failures
that may have been caused by unavailable databases. Prevents end users seeing
abends unnecessarily, enables the application to deal with the situation in
a more user-friendly way, and avoids the need to code global HANDLE ABEND
commands.
- OASN (origin application schedule number)
- An IMS recovery
element in an external subsystem (for example, DB2). The OASN is equivalent to the unit of
recovery ID in the CICS recovery token. It is coupled with the IMS ID to become
the recovery token for UOWs in external subsystems. You can display it using
the DBCTL operation command /DISPLAY and then use the /CHANGE SUBSYS OASN
RESET command to purge incomplete UOWs.
- OLDS (online log data set)
- A data set on direct
access storage that contains the log records written by DBCTL. When the current
OLDS is full, IMS continues logging to a further available OLDS.
- overseer
- A CICS program running in its own address space
that provides status information about active and alternate CICS systems. You
can use it to automate a restart of failed regions.
- PAPL (participant adapter parameter list)
- A component
of the DRA. See also DRA (database resource adapter).
- PAPLRETC
- The response code field from the DRA.
- PCB (program communication block)
- An IMS control block
that describes an application program’s interface to an IMS database or,
additionally, for message processing and batch message processing (BMP) programs,
to the source and destination of messages. See also PSB (program
specification block).
- PDIR (PSB directory)
- Contains entries defining
each PSB to be accessed using local DL/I. Also contains entries for remote
PSBs, to which requests are function-shipped using remote DL/I. A CICS system initialization
parameter of the same name specifies a suffix for the PDIR.
- physical partition
- Part of a central processing
complex (CPC) that operates as a CPC in its own right, with its own copy of
the operating system.
- PI (program isolation)
- An IMS facility that protects all activity of
an application program from any other active application program until that
application program indicates, by reaching a syncpoint, that the data it has
modified is consistent and complete.
- PSB (program specification block)
- An IMS control block
that describes databases and logical message destinations used by an application
program. A PSB consists of one or more program communication blocks (PCBs).
See also PCB (program communication block).
- pseudo recovery token
- A pseudo recovery token consists
of 8 decimal characters, which can be used in place of the recovery token
in certain circumstances. For example, a pseudo recovery token is displayed
when the status of an application thread is in-doubt. It is made shorter so
that it is easier to make note of and enter, for example, in certain DBCTL
commands. See also recovery token.
- PST (partition specification table)
- An IMS control block
that contains information about a dependent region; for example, type of region,
data transferred by DL/I, and status codes. In a CICS-DBCTL environment, the
dependent region is CICS.
- RDS (restart data set)
- An IMS direct access data set used to contain
system checkpoint ID information written during the logging process. The information
is used when restarting IMS (DBCTL). This checkpoint information is contained in
a table called the checkpoint ID table, which contains an entry for each checkpoint
taken. During restart, IMS uses the table to determine from which checkpoint restart
should take place. IMS finds the information it needs and uses it automatically.
If the RDS is not available at restart, you can obtain the checkpoint information
needed from the log, but this may lengthen the restart process.
- (RECON) recovery control data sets
- DBRC automatically
records information in dual recovery control (RECON) data sets. Both data
sets contain identical information, and so are usually referred to as one--the
RECON. You need the information from the RECON during warm and emergency restarts.
DBRC selects the correct data sets to be used by a recovery utility for you
when you enter a GENJCL command. For a restart, the RECON shows which data
set--the OLDS or the SLDS--contains the most recent log data for
each database data set (DBDS) you have registered with DBRC. For the OLDS,
the RECON shows whether the OLDS has been closed and whether it has been archived.
The RECON contains timestamp information for each log data set and volume.
You use this information to determine which data set and volume contain the
checkpoint information needed to restart DBCTL.
- recovery token
- A recovery token is a 16-byte unique
identifier that is created by CICS (and passed to DBCTL) for each UOW. Its
lifetime is the same as the UOW. The first 8 bytes are the CICS APPLID (in
an XRF environment, this is the generic APPLID) and the second 8 bytes are
a unit of recovery ID. (CICS creates a unit of recovery ID for every UOW.) DBCTL
validates the recovery token to protect against duplication of units of recovery.
The DBCTL operator can display the recovery token by using the /DISP CCTL
command. It is also displayed in a number of CICS and IMS messages. See also pseudo
recovery token.
- redo
- A DEDB term, which is more or less analogous
to the full function DL/I term COMMIT. It has the same aim, but the means
of achieving it are different. For DEDBs, if phase two action is COMMIT, the
changes must be written to the database using REDO, because the DEDB changes
have only been made in main storage.
REDO is also used to refer to the
action required for committed DEDBs during emergency restart of IMS. You can determine
from the log that a COMMIT was initiated, but that phase two is not indicated
as complete. In this case, DEDB updates must be REDOne.
- remote DL/I
- Accessing a DL/I database by function
shipping, in which CICS sends a DL/I request to another CICS system. See
also function shipping.
- return code equate
- An alphameric equivalent of
a numeric return code, such as UERCNOAC for "take no action". In DBCTL,
return code equates are used in the XRF global user exits XXDFA, XXDFB, and
XXDTO.
- RMI (resource manager interface)
- A program or a
group of programs that enable you to structure calls from your CICS system in such
a way that they can access non-CICS resources, such as databases, that you
would not normally be able to access. An RMI is written using the CICS task-related
user exit interface. DBCTL is accessed by means of a CICS-supplied RMI. See
also task-related user exit.
- RMODE (residency mode)
- Specifies where a program
is expected to reside in virtual storage. RMODE 24 indicates that a program
is coded to reside in virtual storage below the 16MB line. RMODE ANY indicates
that a program is coded to reside anywhere in virtual storage.
- RIS (recoverable in-doubt structure)
- When a failure
occurs, an RIS is constructed for each unit of recovery and is written to
the IMS log. Its contents include the recovery token, the changed data records,
and the identity of the data block that cannot be accessed because of unresolved
in-doubts.
- RSE (recoverable service element)
- Each recoverable
service element (RSE) contains a set of DBCTL subsystem identifiers of equivalent
DBCTL subsystems together with their associated job names, and the specific
APPLIDs of the CICS systems that will use them.
When CICS attempts to
connect to DBCTL using a particular startup table, it attempts to connect
using the specific subsystem ID associated with that startup table, or any
other DBCTL subsystem ID in the RSE to which the specific subsystem ID belongs.
See also equivalent and RST (recoverable
service table).
- RST (recoverable service table)
- A suffixable table,
specified by a CICS system initialization parameter. You use the CICS RST to define
the relationships between your DBCTL subsystems (DBCTL is a subsystem to MVS)
and CICS systems. The RST consists of a set of recoverable service elements
(RSEs).
CICS can use the RST when connecting to a DBCTL system, or when a connection
to a DBCTL system fails. A CICS alternate can also use the RST to determine
whether it is authorized to cancel a particular DBCTL subsystem. See also equivalent and RSE (recoverable service
element).
- SCHEDULE PSB
- An application schedules a PSB to
obtain access to PCBs. See also PCB (program communication
block) and PSB (program specification block).
- scheduling
- Selecting jobs or tasks that are to
be run.
- single-MVS environment
- An environment that supports
one MVS image. See also MVS image.
- single-phase commit
- A
one-step process by which recoverable resources in an IMS system and a CICS system are committed. CICS can use single-phase commit instead of
two-phase commit when, in a particular LUW, DBCTL is the only recoverable
resource used. Using single-phase commit avoids unnecessary log records being
written, decreases transaction cost and improves response time in both CICS and DBCTL. See also two-phase commit.
- SIT (system initialization table)
- A CICS table that
contains user-specified information to initialize and control system functions,
module suffixes for selection of user-specified versions of CICS modules and
tables, and information used to control the initialization process. You can
generate several SITs and then select the one that best meets your current
requirements at initialization time.
- SLDS (system log data set)
- When the OLDS is full,
it is archived as an SLDS. An SLDS can be on DASD or tape. The contents are
used as input to the database recovery process. See also OLDS (online log data set) and WADS (write ahead data
set).
- snap dump
- A snap (or snapshot) dump can be requested
by a task at any time during which that task is being processed.
- startup job stream
- A set of job control statements
used to initialize CICS.
- statistics
- In CICS and IMS, data produced on timing and resources
used by the system as a whole over a specified period of time. Note that CICS distinguishes between monitoring and statistics, but IMS does not. See
also monitoring.
- subsystem
- A secondary or subordinate system of
the main system; for example, DBCTL, which is a subsystem to MVS.
- SVC (supervisor call)
- An instruction that interrupts
the program being executed and passes control to the supervisor so that it
can perform a specific service indicated by the instruction.
- syncpoint
- A syncpoint (or synchronization point)
is a logical point in execution of an application program where the changes
made to the databases by the program are consistent and complete and can be
committed to the database. The output, which has been held up to that point,
is sent to its destination(s), the input is removed from the message queues,
and the database updates are made available to other applications. CICS recovery and
restart facilities do not backout updates prior to a syncpoint if the program
has terminated abnormally.
A syncpoint is created by any of the following:
- A DL/I CHECKPOINT command or CHKP call
- A DL/I TERMINATE command or TERM call
- A CICS syncpoint request
- An end of task or an end of program.
See also UOW (unit of work).
- SYSPLEX (systems complex)
- In an MVS/ESA environment,
a set of one or more MVS systems given an XCF name and in which programs in
these systems can then use XCF services.
- takeover
- In XRF, the shift of workload from the
active to the alternate CICS system, and the switching of resources needed for
this to happen.
- task-related user exit
- A task-related user exit
enables you to write a user exit program that is associated with specified
events in a particular task, rather than with every occurrence of a particular
event in CICS processing (as is the case with global user exits). Task-related
user exits can be used to build a resource manager interface (RMI) that enables
you to access non-CICS resources, such as databases. DBCTL is accessed by
means of a CICS-supplied RMI. See also global user exit and RMI (resource manager interface).
- TCB (task control block)
- An MVS concept. Anything
in the operating system needs a TCB to execute. In a non-DBCTL environment, CICS needs only one TCB. DBCTL provides a separate TCB for each CICS application
thread, which significantly improves the amount of concurrent processing.
See also thread.
- thread
- A CICS application thread provides a two-way
link between an application and DBCTL. It is the representation in DBCTL of
a CICS transaction when that transaction issues a DL/I request to DBCTL.
The DRA creates a thread for each transaction when it first schedules a PSB.
The thread is terminated and made available for other work at syncpoint or
when an abend occurs. It identifies the transaction’s existence, traces
its progress, sets aside the resources it needs to be processed, and delimits
its accessibility to other resources. You can display and stop threads using IMS operator
commands. You can use them in problem determination and in performance tuning,
because they are displayed (as recovery tokens) in various messages, traces,
dumps, and thread activity is included in DBCTL statistics. See also recovery token and TCB (task control block).
- TOR (terminal-owning region)
- A CICS address space
whose primary purpose is to manage terminals. See also AOR
(application-owning region) and DOR (database-owning
region).
- tracking
- In XRF, monitoring of terminals in the
active CICS system by the alternate CICS system.
- two-phase commit
- A two-step process by which recoverable
resources in an IMS system and a CICS system are committed. During the first
step, the subsystems are polled to ensure that they are ready to commit. If
they all respond positively, they are then instructed to execute commit processing.
- UIB (user interface block)
- Whenever you issue an
DL/I call, DL/I responds by storing the information in the UIB in your program.
Include the UIB in your application program only if it is to be referenced.
The UIB is acquired by the interface routine when an application issues a
schedule request specifying a pointer reference to be set with the address
of the UIB. Information on the success (or otherwise) is returned to UIBFCTR
and UIBDLTR.
- undo
- A DEDB term, which is more or less analogous
to the full function DL/I term BACKOUT. It has the same aim, but the means
of achieving it are different. The difference is in the stage at which updates
are written to the database. For DEDBs, if phase two action of two-phase commit
is ABORT, no changes have to be made to the database, because the changes
are still in main storage, and can be UNDOne from there.
- unit of recovery
- In DBCTL, a unit of recovery is
created for each processing request when the first DL/I update request is
received from CICS and is kept until a two-phase commit is complete.
A unit of recovery is more or less synonymous with a CICS UOW, except
that it begins when the first DL/I request is received from CICS, and not when
the CICS task begins. See UOW (unit of work).
- UOW (unit of work)
- Synonymous with logical unit of work. In CICS, a sequence of processing actions (for
example, database changes) that must be completed before any of the individual
actions can be regarded as committed. A UOW begins when a task starts or at
a syncpoint you specified, and ends at a syncpoint you
specified or when a task ends. If you do not specify any syncpoints, an entire
task will be an UOW.
If changes are committed when the UOW completes successfully
and the syncpoint is recorded on the system log, these changes do not need
to be backed out if there is a subsequent failure of the task or system. See
also unit of recovery and syncpoint.
- user-replaceable program
- A CICS-supplied program
that is always invoked at a particular point in CICS processing as if it were part of CICS code. The program contains points at which you can enter your own code.
DFHDBUEX is a user-replaceable program for use with DBCTL.
- WADS (write ahead data set)
- A data set that contains
log records that reflect committed operations but are not yet written to an OLDS (online log data set).
- XCF (Cross Systems Coupling Facility)
- A facility
of MVS/ESA that provides some initial MVS services needed to support a multisystem
environment while still maintaining a single system image. Systems coupled
using XCF are known as an XCF SYSPLEX.
- XRF (extended recovery facility)
- A software function
that minimizes the impact of various system failures on users by transferring
activity to an alternate system in the same MVS image or a different one.
[[