GGSCI Command    Summary
SUBDIRS          CREATE SUBDIRS
ER               INFO ER, KILL ER, LAG ER, SEND ER, STATUS ER, 
                 START ER, STATS ER, STOP ER
EXTRACT          ADD, ALTER, CLEANUP, DELETE, INFO, KILL,
                 LAG, SEND, START, STATS, STATUS, STOP
EXTTRAIL         ADD, ALTER, DELETE, INFO
GGSEVT           VIEW
MANAGER          INFO, REFRESH, SEND, START, STOP, STATUS
MARKER           INFO
PARAMS           EDIT, VIEW
REPLICAT         ADD, ALTER, CLEANUP, DELETE, INFO, KILL,
                 LAG, SEND, START, STATS, STATUS, STOP
REPORT           VIEW
RMTTRAIL         ADD, ALTER, DELETE, INFO
TRACETABLE       ADD, DELETE, INFO
TRANDATA         ADD, DELETE, INFO
Database         DBLOGIN, LIST TABLES, 
                 ENCRYPT PASSWORD
DDL              DUMPDDL
CHECKPOINTTABLE  ADD CHECKPOINTTABLE, DELETE CHECKPOINTTABLE, 
                 CLEANUP CHECKPOINTTABLE, INFO CHECKPOINTTABLE
Miscellaneous    FC, HELP, HISTORY, INFO ALL, INFO MARKER, OBEY,  
                 SET, SHELL, SHOW, VERSIONS, ! 
              

---------------------------------------------------------------------
CREATE SUBDIRS

Use CREATE SUBDIRS to create the default directories within the 
GoldenGate home directory. Use CREATE SUBDIRS before any other 
configuration tasks.

Syntax:
CREATE SUBDIRS

--------------------------------------------------------------------
INFO ER

Use INFO ER to view the following information for multiple Extract
and Replicat groups as a unit.

* Status of processing (STARTING, RUNNING, STOPPED, ABENDED).
* Checkpoint information.
* Process run history
* Trail information.
* Approximate time lag and byte lag.

The basic INFO ER command, without either the TASKS or 
ALLPROCESSES argument, displays information only for online 
(continuous) groups. Tasks are excluded. 

Syntax:
INFO ER <group spec> [, <option>] [, ...]

<group spec>
A group name or a wildcard (*) to specify multiple groups.

[<option>] can be one of the following:

SHOWCH [<n>]
The basic command shows information about the current Extract
checkpoints. Specify a value for <n> to include the specified number 
of previous checkpoints as well as the current one.

DETAIL
Displays run history (including start and stop points in the 
data source) and trails.

TASKS
Displays only tasks. Task types that were specified by a
wildcard argument are not displayed by INFO ER. 

ALLPROCESSES
Displays online groups and tasks.

Examples:
INFO ER fin*, SHOWCH
INFO ER *, TASKS

--------------------------------------------------------------------
KILL ER

Use KILL ER to kill multiple Extract and Replicat groups as a unit.
Use this command only if processes cannot be stopped gracefully with
the STOP ER command. The Manager process will not attempt to restart
a killed process.

Syntax:
KILL ER <group spec>

<group spec>
The name of a group or a wildcard (*) to specify multiple groups.

Example: KILL ER f*

---------------------------------------------------------------------
LAG ER

Use LAG ER to determine a true lag time for multiple Extract and
Replicat groups as a unit. Lag time is the difference in seconds
between the time a record was processed and the timestamp of that
record in the data source (transaction log, log table, or trail).
LAG ER calculates the lag time more precisely than INFO ER because it
communicates with a process directly rather than reading a checkpoint
position.

Syntax:
LAG ER <group spec>

<group spec>
The name of a group or a wildcard (*) to specify multiple groups. 

Example 1: LAG ER *
Example 2: LAG ER *FIN*

--------------------------------------------------------------------
SEND ER

Use SEND ER to communicate with multiple running Extract and
Replicat groups as a unit. The request is processed as soon as the
process is ready to accept commands from users.

Syntax:
SEND ER <group spec>, <command>

<group spec>
The name of a group or a wildcard to specify multiple groups. If a
group is not running, an error is returned.

<command> can be one of the following:

FORCESTOP
Forces the processes to stop immediately.

GETLAG
Determines a true lag time between the process and the data source. 
Returns the same results as LAG ER.

REPORT
Generates an interim statistical report to the Extract report file. 
The statistics that are displayed depend upon the configuration of 
the STATSOPTIONS parameter when used with the RESETREPORTSTATS |
NORESETREPORTSTATS option. DDL information is also included. 

STATUS
Returns the current location within the data source.

STOP
Stops the processes.

TRACE[2] <tracefile> | OFF
Turns tracing on and off. Tracing captures information to the 
specified file to reveal processing bottlenecks. TRACE captures step-
by-step processing information. TRACE2 identifies code segments 
rather than specific steps. If a trace is already  running, the
associated trace file is closed and the trace is resumed with output
to the new file specified with <tracefile>. Contact GoldenGate
Technical Support for assistance if the trace reveals significant 
processing bottlenecks. Turn off tracing with TRACE OFF or TRACE2 OFF.

TRACEINIT
Initializes (resets) tracing statistics back to 0 and then starts
accumulating statistics again. Use this option to track the current
behavior of processing, as opposed to historical.


Example 1: SEND ER fi*, REPORT
Example 2: SEND ER fi*, STOP

--------------------------------------------------------------------
STATUS ER

Use STATUS ER to determine whether or not multiple Extract and
Replicat processes are running.

Syntax:	
STATUS ER <group spec> [, <option> ]

<group spec>
The name of a group or a wildcard (*) to specify multiple groups.

[<option>] can be one of the following:
	
TASKS
Displays status only for tasks. By default, tasks are not displayed
unless you specify a single group name (without wildcards).

ALLPROCESSES
Displays status for all groups, including tasks.

Example 1: STATUS ER fin*

--------------------------------------------------------------------
START ER

Use START ER to start multiple Extract and Replicat groups as a unit.
To confirm that processes have started, use the INFO ER or STATUS ER
command.

Syntax:	
START ER <group spec>

<group spec>
The name of a group or a wildcard (*) to specify multiple groups.

Example: START ER fin*

----------------------------------------------------------------------
STATS ER

Use STATS ER to display statistics for multiple Extract and Replicat
groups as a unit.

To get the most accurate number of operations per second that are being
processed, issue the STATS ER command with the RESET option, then wait
a few minutes and issue the STATS ER command with the REPORTRATE
option. The LATEST STATISTICS field shows the operations per second.
 
Syntax:
STATS ER <group spec> [, <option>] [, ...]

<group spec>
The name of a group or a wildcard (*) to specify multiple groups.

[<option>] Can be any of the following:
	
<statistic>
The statistic to be displayed. Valid values are:
   TOTAL: Displays totals since processing startup.
   DAILY: Displays totals since the start of the current day.
   HOURLY: Displays totals since the start of the current hour.
   LATEST: Displays totals since the last RESET command.
   RESET: Resets the counters in the LATEST statistical field.
More than one statistic can be specified by separating each with a
comma, for example STATS ER fin*, TOTAL, DAILY.
	
TABLE <table>
Displays statistics only for the specified table or a group of tables
specified with a wildcard (*).
	
TOTALSONLY <table spec>
Totals statistics for the specified table or a group of tables 
specified with a wildcard (*).
	
REPORTRATE <time  units>
Displays statistics in terms of processing rate rather than absolute 
values. Valid values are HR, MIN, or SEC.

Example: The following example displays total and hourly statistics 
per minute for a table and also resets the latest statistics.

STATS ER fin*, TOTAL, HOURLY, TABLE acct, REPORTRATE MIN, RESET

--------------------------------------------------------------------
STOP ER

Use STOP ER to stop multiple Extract and Replicat groups gracefully as
a unit. The command preserves the state of synchronization for the next
time the process starts, and it ensures that Manager does not start
processes automatically.

Syntax:
STOP ER <group spec>

<group spec>
The name of a group or a wildcard (*) to specify multiple groups.


Example: STOP ER fin*

---------------------------------------------------------------------
####################################
#
#EXTRACT COMMANDS
#
#
####################################

---------------------------------------------------------------------
ADD EXTRACT

Use ADD EXTRACT to create an Extract group. Unless a SOURCEISTABLE
task is specified, this command creates checkpoints so that 
processing continuity is maintained from run to run.

Syntax for regular, passive, or data pump Extract:
ADD EXTRACT <group name>
{, SOURCEISTABLE |
   , GGSLOG <ggslog table> |
   , TRANLOG [bsds name> |
   , VAM |
   , EXTFILESOURCE <file name> |
   , EXTTRAILSOURCE <trail name> |
   , VAMTRAILSOURCE <VAM trail name>}
{, BEGIN <begin spec> | 
   , EXTSEQNO <seqno>, EXTRBA <relative byte address> |
   , LOGNUM <log number>, LOGPOS <byte offset> |
   , LSN <value> |
   , EXTRBA <relative byte address> |
   , EOF | LSN <value> |
   , PAGE <data page>, ROW <row>}
[, THREADS <n>]
[, PASSIVE]
[, PARAMS <parameter file>]
[, REPORT <report file>]
[, DESC <description>]

Syntax for alias Extract:
ADD EXTRACT <group name>
, RMTHOST {<host name> | <IP address>}
, {MGRPORT <port>} | {PORT <port}
[, RMTNAME <name>]
[, DESC <description>]

<group name>
The name of the Extract group. Follow these naming conventions:
* You can use up to eight ASCII characters, including nonalphanumeric
  characters such as the underscore (_). Any ASCII character can be
  used, so long as the operating system allows that character to be
  in a filename. This is because a group is identified by its
  associated checkpoint file.
* Group names are not case-sensitive.
* Use only one word.
* Do not use the word port as a group name. However, you can use
  the string port as part of the group name.
* Do not place a numeric value at the end of a group name, such as
  fin1, fin10, and so forth. You can place a numeric value at the
  beginning of a group name, such as 1_fin, 1fin, and so forth.
   ext_1
   ex+2t
   ex!2t

The following are datasource options:

SOURCEISTABLE
Creates an Extract task that extracts entire records from the 
database for an initial load. If SOURCEISTABLE is not specified,
ADD EXTRACT creates an online change-synchronization process, and one
of the other datasource options must be specified. When using
SOURCEISTABLE, do not specify service options. Task parameters must be
specified in the parameter file.

GGSLOG <ggslog table>
Specifies the GoldenGate log table as the data source. Use this 
option for trigger-based extraction. The default name is GGSLOG.

TRANLOG
Specifies the transaction log as the data source. Use this option 
for log-based extraction. TRANLOG requires either the BEGIN or EXTSEQNO
and EXTRBA options.

Use the <bsds name> option for DB2 on z/OS to specify the
BSDS (Bootstrap Data Set) file name of the transaction log.

VAM
Indicates that a Vendor Access Module will be used to extract data
from the data source.

EXTFILESOURCE <file name>
Specifies an extract file as the data source. Use this option with a
secondary Extract group (data pump) that acts as an intermediary
between a primary Extract group and the target system. For <file name>,
specify the fully qualified path name of the file, for example
c:\ggs\dirdat\extfile.

EXTTRAILSOURCE <trail name>
Specifies a trail as the data source. Use this option with a secondary
Extract group (data pump) that acts as an intermediary between a
primary Extract group and the target system. For <trail name>, specify
the fully qualified path name of the trail, for example
c:\ggs\dirdat\aa.

VAMTRAILSOURCE <VAM trail name>
Specifies a VAM trail. Use this option if using Teradata VAM-based
extraction in maximum protection mode. Specify the fully qualified path
name of the VAM trail to which the primary Extract group is writing.
Use a VAM-sort Extract group to read the VAM trail and send the data to
the target system.

The following are service options:

BEGIN <begin spec>
Specifies a timestamp in the data source at which to begin processing.
Valid values:
* NOW
* A date and time in the format of:
  yyyy-mm-dd [hh:mi:[ss[.cccccc]]]

**What NOW means**

  For all databases except DB2 LUW, NOW specifies the time at which the
  ADD EXTRACT command is issued.

  For DB2 LUW, NOW specifies the time at which START EXTRACT takes
  effect. It positions to the first record that approximately matches
  the date and time. This is because the only log records that contain
  timestamps are the commit and abort transaction records, so the
  starting position can only be calculated relative to those
  timestamps. This is a limitation of the API that is used by
  GoldenGate.

  Do not use NOW for a data pump Extract unless you want to bypass any
  data that was captured to the GoldenGate trail prior to the ADD
  EXTRACT statement.

EXTSEQNO <seqno>, EXTRBA <relative byte address>
Valid for a primary Extract for Oracle and NonStop SQL/MX, and for a
data pump Extract. Specifies either of the following:
* sequence number of an Oracle redo log and RBA within that log at
  which to begin capturing data.
* the NonStop SQL/MX TMF audit trail sequence number and relative byte
  address within that file (location in the TMF Master Audit Trial
  (MAT) at which to begin capturing data.
* the file in a trail in which to begin capturing data (for a data
  pump). Specify the sequence number, not any zeroes used for padding.
  For example, if the trail file is c:\ggs\dirdat\aa000026, you would
  specify EXTSEQNO 26. By default, processing begins at the beginning
  of a trail unless this option is used.
Contact GoldenGate Technical Support before using this option.  

LOGNUM <log number>, LOGPOS <byte offset>
Valid for c-tree. Specifies the location in a c-tree transaction log at
which to start capturing data. 
* <log number> is the number of the c-tree log file. 
* <byte offset> is the relative position from the beginning of the file
  (0 based).

LSN <value>
Valid for SQL Server or Ingres. Specifies the LSN in a SQL Server or
Ingres transaction log at which to start capturing data. The LSN
specified should exist in a log backup or the online log.

*For SQL Server, an LSN is composed of one of these, depending on how
the database returns it:
** Colon separated Hex string (8:8:4) padded with leading zeroes and 
   0X prefix, as in 0X00000d7e:0000036b:01bd 
** Colon separated Decimal string (10:10:5) padded with leading zeroes,
   as in 0000003454:0000000875:00445
** Colon separated Hex string with 0X prefix and without leading
   zeroes, as in 0Xd7e:36b:1bd 
** Colon separated Decimal string without leading zeroes, as in
   3454:875:445 
** Decimal string, as in 3454000000087500445
 
In the preceding, the first value is the virtual log file number, the
second is the segment number within the virtual log, and the third is
the entry number.

*For Ingres, an LSN is two, 4-byte unsigned integers, separated by a
colon. For example, to specify an LSN of 1206396546,43927 (as viewed
in an Ingres utility), you would enter 1206396546:43927.

An alias for this option is EXTLSN. 

EXTRBA <relative byte address> 
Specifies the relative byte address in a DB2 transaction log at which
to start extracting data. Valid only for DB2 on the z/OS platform.

EOF | LSN <value>
Valid for DB2 LUW. Specifies a start position in the DB2 LUW
transaction logs when Extract starts.

* EOF configures processing to start at the active LSN in the log
  files. The active LSN is the position at the end of the log files
  that the next record will be written to. Any active transactions
  will not be captured. 
* LSN <value> configures processing to start at an exact LSN if a
  valid log record exists there. If one does not exist, Extract will
  abend. Note that, while Extract might position to a given LSN,
  that LSN might not necessarily be the first one that Extract will
  process. There are numerous record types in the log files that
  Extract ignores, such as DB2 internal log records. Extract will
  report the actual starting LSN to the Extract report file. 

PAGE <data page>, ROW <row>
Valid for Sybase. Specifies a data page and row that together define a
start position in a Sybase transaction log.

PARAMS <parameter file>
Specifies an Extract parameter file in a location other than the
default of dirprm within the GoldenGate directory. Use the fully
qualified path name.

REPORT <report file>
Specifies an Extract report file in a location other than the default
of dirrpt within the GoldenGate directory. Use the fully qualified
path name.

THREADS <n>
Specifies the number of redo threads when extracting data from an
Oracle RAC clustered database configuration.

PASSIVE
Specifies that this Extract group runs in passive mode and can only be
started and stopped by starting or stopping an alias Extract group on
the target system. Source-target connections will be established not by
this group, but by the alias Extract from the target. PASSIVE can be
used for a regular Extract group or a data-pump Extract group. It
should only be used by whichever Extract on the source system is the
one that will be sending the data across the network to a remote
trail on the target.

DESC <description>
Specifies a description of the group, such as Extracts account_tab on
Serv1. The description must be within quotes. You may use the
abbreviated keyword DESC or the full word DESCRIPTION.

RMTHOST {<host name> | <IP address>}
Identifies this group as an alias Extract and specifies either the DNS
name of the remote host or its IP address.

{MGRPORT <port>} | {PORT <port}
Use this option for an alias Extract. Use one of the following:
  MGRPORT specifies the port on the remote system where Manager is
  running. Use this option when using a dynamic Collector.
  PORT specifies a static Collector port. Use instead of MGRPORT 
  only if running a static Collector.

RMTNAME <name>
Use for an alias Extract. Specifies the passive Extract name,
if different from that of the alias Extract.

Example 1:
The following creates an Extract group named finance that extracts
database changes from the beginning of the GoldenGate log table named
chglog. 

ADD EXTRACT finance, GGSLOG chglog

Example 2:
The following creates an Extract group named finance that extracts
database changes from the transaction logs. Extraction starts with
records generated at the time when the group was created.

ADD EXTRACT finance, TRANLOG, BEGIN NOW

Example 3:
The following creates an Extract group named finance that extracts 
database changes from Oracle RAC logs. Extraction starts with records
generated at the time when the group was created. There are four RAC
instances, meaning there will be four Extract threads.

ADD EXTRACT finance, TRANLOG, BEGIN NOW, THREADS 4

Example 4:
The following creates an Extract group named finance that extracts
database changes from the transaction logs. Extraction starts with
records generated at 8:00 on January 31, 2006.

ADD EXTRACT finance, TRANLOG, BEGIN 2006-01-31 08:00

Example 5:
The following creates an Extract group named finance that interfaces
with a VAM in either maximum performance or maximum protection mode. No
BEGIN point is used for Teradata sources.

ADD EXTRACT finance, VAM

Example 6:
The following creates a VAM-sort Extract group named finance. 
The process reads from the VAM trail /ggs/dirdat/vt.

ADD EXTRACT finance, VAMTRAILSOURCE /ggs/dirdat/vt

Example 7:
The following creates a data-pump Extract group named finance.
It reads from the GoldenGate trail c:\ggs\dirdat\lt.

ADD EXTRACT finance, EXTTRAILSOURCE c:\ggs\dirdat\lt

Example 8:
The following creates an initial-load Extract named load.

ADD EXTRACT load, SOURCEISTABLE

Example 9:
The following creates a passive Extract group named finance
that extracts database changes from the transaction logs.
 
ADD EXTRACT finance, TRANLOG, BEGIN NOW, PASSIVE

Example 10:
The following creates an alias Extract group named financeA.
The alias Extract is associated with a passive extract named 
finance on source system sysA. The Manager on that system is
using port 7800.

ADD EXTRACT financeA, RMTHOST sysA, MGRPORT 7800, RMTNAME finance

---------------------------------------------------------------------
ALTER EXTRACT

Use ALTER EXTRACT to change the attributes of an Extract group 
specified with the ADD EXTRACT command, or to increment the trail to
the next file and, optionally, purge the existing files. 

The following are exceptions. For these exceptions, GoldenGate
recommends deleting the Extract group and then adding it again.

* Altering a trail specified with the EXTTRAILSOURCE option of ADD
  EXTRACT.

* Altering the number of RAC threads (THREADS <n>). After deleting
  The Extract group, add it again with the new number of threads.

Before using this command, stop Extract by issuing the STOP EXTRACT 
<group name> command.

Syntax:
ALTER EXTRACT <group name> [, <option>] 
[, ETROLLOVER]
[, THREAD <number>] 
[, ...]

<group name>	
The name of an Extract group.

[<option>] can be any of the ADD EXTRACT options, except you cannot
alter the number of RAC redo threads (THREADS <n>). You must delete
the Extract group and then add it with the new number of threads. 

SOURCEISTABLE
Creates an Extract task that extracts entire records from the 
database for an initial load. If SOURCEISTABLE is not specified,
the command creates an online change-synchronization process, and a
datasource must be specified. When using SOURCEISTABLE, do not 
specify service options. Task parameters must be specified in the
parameter file.

The following are datasource options:

GGSLOG <ggslog table>
Specifies the GoldenGate log table as the data source. Use this 
option for trigger-based extraction. The default name is GGSLOG.

TRANLOG [<bsds name>]
Specifies the transaction logs as the data source. Use this option 
for log-based extraction. Use the <bsds> option for DB2 on z/OS 
to specify the BSDS (Bootstrap Data Set) file name of the
transaction log. TRANLOG requires either the BEGIN or EXTSEQNO and
EXTRBA options.

EXTFILESOURCE <file name>
Specifies an extract file as the data source. Use this option with a
secondary Extract group (data pump) that acts as an intermediary
between a primary Extract group and the target system. Can be used for
either trigger-based or log-based extraction. For <file name>, specify
the fully qualified path name of the file, for example
c:\ggs\dirdat\extfile.

EXTTRAILSOURCE <trail name>
Specifies a trail as the data source. Use this option with a secondary
Extract group (data pump) that acts as an intermediary between a
primary Extract group and the target system. Can be used for either
trigger-based or log-based extraction. For <trail name>, specify the
fully qualified path name of the trail, for example c:\ggs\dirdat\aa.

Note: You can change the source trail for an Extract group with ALTER
EXTRACT, but GoldenGate recommends deleting the group, then adding it
again instead.

VAM
Indicates that a Vendor Access Module will be used to extract data
from the data source.

VAMTRAILSOURCE <VAM trail name>
Specifies a VAM trail. Use this option if using Teradata VAM-based
extraction in maximum protection mode. Specify the fully qualified path
name of the VAM trail to which the primary Extract group is writing.
Use a VAM-sort Extract group to read the VAM trail and send the data to
the target system.

The following are service options:

BEGIN <begin spec>
Specifies a timestamp in the data source at which to begin processing.
Valid values:
* NOW
* A date and time in the format of:
  yyyy-mm-dd [hh:mi:[ss[.cccccc]]]

**What NOW means**

  For all databases except DB2 LUW, NOW specifies the time at which the
  group is created with ADD EXTRACT.

  For DB2 LUW, NOW specifies the time at which START EXTRACT takes
  effect. It positions to the first record that approximately matches
  the date and time. This is because the only log records that contain
  timestamps are the commit and abort transaction records, so the
  starting position can only be calculated relative to those
  timestamps. This is a limitation of the API that is used by
  GoldenGate.

  Do not use NOW for a data pump Extract unless you want to bypass any
  data that was captured to the GoldenGate trail prior to the ADD
  EXTRACT statement.

EXTSEQNO <seqno>, EXTRBA <relative byte address>
Specifies either of the following:
* sequence number of an Oracle redo log and RBA within that log at
  which to begin processing (for a primary Extract).
* the NonStop SQL/MX TMF audit trail sequence number and relative byte
  address within that file (location in the TMF Master Audit Trial
  (MAT).
* * the file in a trail in which to begin processing (for a data pump).
  Specify the sequence number, not any zeroes used for padding. For
  example, if the trail file is c:\ggs\dirdat\aa000026, you would
  specify EXTSEQNO 26. By default, processing begins at the beginning
  of a trail unless this option is used.
Contact GoldenGate Technical Support before using this option.  

LOGNUM <log number>, LOGPOS <byte offset>
Specifies the location in a c-tree transaction log at which to start
processing. 
* <log number> is the number of the c-tree log file. 
* <byte offset> is the relative position from the beginning of the file
  (0 based).

LSN <value>
Valid for SQL Server or Ingres. Specifies the LSN in a SQL Server or
Ingres transaction log at which to start capturing data. The LSN
specified should exist in a log backup or the online log.

*For SQL Server, an LSN is composed of one of these, depending on how
the database returns it:
** Colon separated Hex string (8:8:4) padded with leading zeroes and 
   0X prefix, as in 0X00000d7e:0000036b:01bd 
** Colon separated Decimal string (10:10:5) padded with leading zeroes,
   as in 0000003454:0000000875:00445
** Colon separated Hex string with 0X prefix and without leading
   zeroes, as in 0Xd7e:36b:1bd 
** Colon separated Decimal string without leading zeroes, as in
   3454:875:445 
** Decimal string, as in 3454000000087500445
 
In the preceding, the first value is the virtual log file number, the
second is the segment number within the virtual log, and the third is
the entry number.

 *For Ingres, an LSN is two, 4-byte unsigned integers, separated by a
 colon. For example, to specify an LSN of 1206396546,43927 (as viewed
in an Ingres utility), you would enter 1206396546:43927.

An alias for this option is EXTLSN. 

EXTRBA <relative byte address> 
Specifies the relative byte address in a DB2 transaction log at which 
to start extracting data. Valid for DB2 on the z/OS platform only.

EOF
Specifies processing to start at the end of the transaction logs. Any
active transactions will not be captured. Use this option only for DB2
LUW (Linux, UNIX, Windows).

PAGE <data page>, ROW <row>
Specifies a data page and row that together define a start position in 
a Sybase transaction log.

Other ALTER options:

PARAMS <parameter file>
Specifies an Extract parameter file in a location other than the
default of dirprm within the GoldenGate directory. Use the fully
qualified path name.

REPORT <report file>
Specifies an Extract report file in a location other than the default
of dirrpt within the GoldenGate directory. Use the fully qualified
path name.

DESC <description>
Specifies a description of the group, such as Extracts account_tab on
Serv1. The description must be within quotes. You may use the
abbreviated keyword DESC or the full word DESCRIPTION.

ETROLLOVER
Causes the Extract group to increment to the next file in the trail
when restarting. For example, if the current file is ET000002, the
current file will be ET000003 when the Extract group restarts.

THREAD <number>
In an Oracle RAC configuration, alters Extract only for the specified
redo thread. Only one thread number can be specified.

Example 1: ALTER EXTRACT fin, BEGIN 2006-01-01, ETROLLOVER, ETPURGE
Example 2: ALTER EXTRACT accounts, THREAD 4, BEGIN YYYY-MM-DD
Example 3: ALTER EXTRACT sales, LSN 0X00000d7e:0000036b:01bd 

---------------------------------------------------------------------
CLEANUP EXTRACT

Use CLEANUP EXTRACT to delete run history for a specified Extract 
group. The cleanup keeps the last run record intact so that the group 
can resume processing from where it left off. Before using this
command, stop Extract by issuing the STOP EXTRACT <group name> 
command.

Syntax:	
CLEANUP EXTRACT <group name> [, SAVE <count>]

<group name>
An Extract group name or a wildcard (*) to specify multiple groups. 
For example, T* cleans up all Extract groups whose names start
with T.

SAVE <count>
Excludes the specified number of the most recent records from the 
cleanup.

Example 1: The following deletes all but the last record.

CLEANUP EXTRACT finance

Example 2: The following deletes all but the most recent five records

CLEANUP EXTRACT *, SAVE 5

---------------------------------------------------------------------
DELETE EXTRACT

Use DELETE EXTRACT to delete an Extract group. To delete associated
trail files, delete them manually through the operating system. 
Before using this command, stop Extract by issuing the STOP EXTRACT
<group name> command.

Syntax:	
DELETE EXTRACT <group name> [!]

<group name>
The name of an Extract group or a wildcard specification (*) to
specify multiple groups. For example, T* deletes all Extract groups
whose names start with T.

!
(Exclamation point) Deletes all Extract groups associated with a 
wildcard without prompting. 

Example: DELETE EXTRACT fin* !

---------------------------------------------------------------------
INFO EXTRACT

Use INFO EXTRACT to view the following information.
* Status of Extract processing (STARTING, RUNNING, STOPPED, ABENDED).
* Checkpoint information.
* Process run history
* The trail(s) to which The Extract group is writing.
* Approximate Extract time lag and byte lag.

Time lag reflects the lag in seconds at the time the last checkpoint 
was written to the trail. For example, if the current time is
15:00:00, and the last checkpoint was at 14:59:00, and the timestamp 
of the last record processed was 14:58:00, then the lag is reported as
00:01:00 (one minute, the difference between 14:58 and 14:59). A lag
value of UNKNOWN indicates that Extract might have started but has not
yet processed records, or that the source system's clock is ahead of
the target system's clock (due to clock imperfections, not time zone 
differences). For more precise lag information, use LAG EXTRACT.

The basic INFO EXTRACT command, without either the TASKS or 
ALLPROCESSES argument, displays information only for online 
(continuous) Extract groups. Tasks are excluded. 

Syntax:
INFO EXTRACT <group name> 
[, SHOWCH [<n>]] 
[, DETAIL]
[, TASKS | ALLPROCESSES]
[, ...]

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* returns information for all Extract groups
whose names start with T.

SHOWCH [<n>]
The basic command shows information about the current Extract
checkpoints. Specify a value for <n> to include the specified number 
of previous checkpoints as well as the current one.

DETAIL
Displays Extract run history (including start and stop points in the 
data source) and trails to which the Extract group is writing.

TASKS
Displays only Extract tasks. Task types that were specified by a
wildcard argument are not displayed by INFO EXTRACT. 

ALLPROCESSES
Displays all Extract groups, including tasks.

Examples:
INFO EXTRACT fin*, SHOWCH
INFO EXTRACT *, TASKS

---------------------------------------------------------------------
KILL EXTRACT

Use KILL EXTRACT to kill an Extract process running in regular or
passive mode. Use this command only if Extract cannot be stopped
gracefully with the STOP EXTRACT command. The Manager process will
not attempt to restart a killed Extract process.

Syntax:
KILL EXTRACT <group name>

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* kills all Extract groups whose names start
with T.

Example: KILL EXTRACT finance

---------------------------------------------------------------------
LAG EXTRACT

Use LAG EXTRACT to determine a true lag time between Extract
processing and the data source. Lag time is the difference in seconds
between the time a record was processed by Extract and the timestamp 
of that record in the data source. LAG EXTRACT calculates the lag time
more precisely than INFO EXTRACT because it communicates with the
Extract process directly rather than reading a checkpoint position in
the trail.

Syntax:
LAG EXTRACT <group name>

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* determines lag time for all Extract groups
whose names start with T. 

Example 1: LAG EXTRACT *
Example 2: LAG EXTRACT *FIN*

---------------------------------------------------------------------
SEND EXTRACT

Use SEND EXTRACT to communicate with a running Extract process.
The request is processed as soon as the process is ready to accept
commands from users.

Syntax:
SEND EXTRACT <group name>,
CACHEMGR {CACHESTATS | CACHEQUEUES | CACHEPOOL} |
FORCESTOP |
FORCETRANS <ID> [THREAD <n>] [FORCE] |
GETLAG | 
LOGEND | 
REPORT | 
ROLLOVER | 
SHOWTRANS [<ID>] [THREAD <n>] [COUNT <n>] [DURATION <duration><unit>]
   [TABULAR] [FILE <name> [DETAIL]] |
SKIPTRANS <ID> [THREAD <n>] [FORCE] |
STATUS | 
STOP |
TLTRACE {DEBUG | OFF | <level>} [DDLINCLUDE | DDL[ONLY]] 
   [FILE] <file name> |
TRACE[2] <tracefile> |
TRACE[2] OFF |
TRACEINIT | 
TRANLOGOPTIONS {PURGEORPHANEDTRANSACTIONS |
   NOPURGEORPHANEDTRANSACTIONS} |
TRANLOGOPTIONS TRANSCLEANUPFREQUENCY <minutes> |
VAMMESSAGE <Teradata command> |
VAMMESSAGE {ARSTATS | INCLUDELIST [filter] | 
   EXCLUDELIST [filter]} |
VAMMESSAGE OPENTRANS


<group name>
The name of an Extract group. If the group is not running, an error
is returned.

CACHEMGR {CACHESTATS | CACHEQUEUES | CACHEPOOL}
Retrieves statistics about the GoldenGate memory cache
manager.

* CACHESTATS returns statistics about the overall utilization
  of virtual memory, such as that which is being used, that
  which has been recycled for possible reassignment, that
  which is in anonymous memory, and that which
  corresponds to paged virtual memory. Also provided are
  the size of the cache and the thresholds related to buffer
  allocation and paging, and statistics related to the cache
  files on disk and their related I/O.

* CACHEQUEUES returns statistics about the queues, which
  are memory buffers that have been recycled for future
  use. Of particular interest are a statistic that shows the
  number of times a buffer of the required size was found,
  and the number of times that such a buffer could not be
  found, resulting in a new request for more virtual
  memory.

* CACHEPOOL returns statistics about the cache sub-pools,
  which consist of one sub-pool per log reader thread ( one
  or more in Oracle RAC; one otherwise) for most
  transaction row data, and another sub-pool for BLOB
  data and possibly other large items. The actual allocation
  of data to one or the other of these pool types might not
  always correspond to the type of data that is being stored.
 
These statistics can be used as the basis for adjusting the
CACHEMGR parameter. However, there is generally no need to
view these statistics or change CACHEMGR because the cache
manager is self-adjusting. Do not make any adjustments to
the memory cache manager without the guidance of a
GoldenGate support analyst.

FORCESTOP
Forces Extract to stop, bypassing any notifications. This command will
stop the process immediately.

FORCETRANS <ID> [THREAD <n>] [FORCE]
Forces Extract to write the specified transaction to the trail. Get the
ID number with SHOWTRANS or from an Extract error message. Any data 
added to this transaction afterward is ignored.

* Use THREAD <n> to specify which thread generated the transaction in 
  a RAC environment if there are duplicate transaction IDs. 

* Use FORCE to bypass the confirmation prompt. 

FORCETRANS does not commit the transaction to the source database. It 
only forces the existing data to the trail so that it is processed 
(with an implicit commit) by Replicat. In order to use FORCETRANS, the 
transaction specified must be the oldest one in the list of open 
transactions shown with SHOWTRANS. You can repeat the command for other
transactions in order of their age.

After using FORCETRANS, wait at least five minutes before issuing a
SEND EXTRACT with FORCESTOP. Otherwise, the transaction will still be
present.

If FORCETRANS is used immediately after Extract starts, you may see the
following message:
  This transaction cannot be forced to extract trail at this moment.
  Please try again in a few minutes.
This means that no other transactions have been processed yet by
Extract. Once another transaction is processed, you will be able to
force the transaction to trail.

GETLAG
Determines a true lag time between Extract and the data source. 
Returns the same results as LAG EXTRACT.

LOGEND
Confirms whether or not the Extract process has processed all of 
the records in the data source.

REPORT
Generates an interim statistical report to the Extract report file. 
The statistics that are displayed depend upon the configuration of
the STATOPTIONS parameter when used with the RESETREPORTSTATS |
NORESETREPORTSTATS option.
	
ROLLOVER
Causes Extract to increment to the next file in the trail when
restarting. For example, if the current file is ET000002, the current
file will be ET000003 after the command executes.

SHOWTRANS
Displays information about open transactions. Without options, displays 
all open transactions that will fit into the 25K buffer.

* <ID> constrains the output to a specific transaction based on its 
  ID number.

* THREAD <n> constrains the output to open transactions against a 
  specific Oracle RAC thread. For <n>, use a thread number recognized 
  by Extract.

* COUNT <n> constrains the output to the specified number of 
  open transactions, starting with the oldest one.

* DURATION <duration><unit> restricts the output to transactions that 
  have been open longer than the specified time, where:

   <duration> is the length of time expressed as a whole number.
   <unit> is seconds, minutes, hours, or days in fully spelled out 
   or abbreviated form:

   S|SEC|SECS|SECOND|SECONDS
   M|MIN|MINS|MINUTE|MINUTES
   H|HOUR|HOURS
   D|DAY|DAYS

* TABULAR generates output in tabular format similar to the default
  table printout from SQL*Plus. The default is field-per-row.

* FILE <name> [DETAIL] forces Extract to write the transaction
  information to the specified file. There is no output to the console.
  To write a hex and plain character dump of the data, use FILE with
  DETAIL. This dumps the entire transaction from memory to the file.
  Viewing the data may help you decide whether to skip the transaction
  or force it to the trail. Note: Basic detail information is 
  automatically written to the report file at intervals specified by 
  the WARNLONGTRANS CHECKINTERVAL parameter.

SKIPTRANS <ID> [THREAD <n>] [FORCE]
Forces Extract to skip the specified transaction. Get the ID number
with SHOWTRANS or from an Extract error message. 

* Use THREAD <n> to specify which thread generated the transaction in a 
  RAC environment if there are duplicate transaction IDs. 

* Use FORCE to bypass the confirmation prompt.

Note: In order to use SKIPTRANS, the transaction specified must be the
oldest one in the list of open transactions shown with SHOWTRANS. You
can repeat the command for other transactions in order of their age.

After using SKIPTRANS, wait at least five minutes before issuing a
SEND EXTRACT with FORCESTOP. Otherwise, the transaction will still be
present.

STATUS
Returns a detailed status of the processing state, including current
position and activity. Possible processing status messages are:
* Delaying  waiting for more data
* Processing data  processing data
* Starting initial load  starting an initial load task
* Processing source tables  processing data for initial load task
* Reading from data source  reading from either the source table, 
  the GoldenGate log table, the transaction log, or the VAM interface
* Adding record to transaction list  adding a record to the file
  memory transaction list
* At EOF (end of file)  no more records to process
The command also returns the group name and process ID, the position in
the data source where applicable, timestamp information, and other
relevant information, depending on the database type.

STOP
Stops Extract. If there are any long-running transactions, a message is 
displayed prompting you to make the archives containing data for those 
transactions available for when Extract restarts. To force Extract to 
stop, use the SEND EXTRACT <group>, FORCESTOP command.

TLTRACE {DEBUG | OFF | <level>} [DDLINCLUDE | DDL[ONLY]] 
   [FILE] <file name> 
(Valid for Oracle) Turns detailed tracing of Oracle transaction log
activity on or off. 

 *  The <level> option specifies the detail of the trace, expressed as
    an integer between 1 and 99. Trace information is sent to the
    report file. This option is for debugging and should be used with
    guidance from GoldenGate Technical Support.

 *  DDLINCLUDE | DDLONLY enables DDL tracing and
    specifies how DDL tracing is included in the trace report.
    DDLINCLUDE includes DDL tracing in addition to regular tracing of
    transactional data processing. DDL[ONLY] excludes the tracing of
    transactional data processing and only traces DDL. This option can
    be abbreviated to DDL.

  * [FILE] <file name> specifies the fully qualified name of a file to
    which GoldenGate logs the trace information. The FILE keyword is
    optional, but must be used if other parameter options will follow
    the file name, for example:
    SEND EXTRACT <group> TRACE FILE <file name> DDLINCLUDE
    If no other options will follow the file name, the FILE keyword can
    be omitted, for example:
    SEND EXTRACT <group> TRACE DDLINCLUDE <file name>

TRACE[2] <tracefile> | TRACE[2] OFF
Turns tracing on and off. Tracing captures information to the 
specified file to reveal processing bottlenecks. TRACE captures step-
by-step processing information. TRACE2 identifies code segments 
rather than specific steps. If a trace is already  running, the
associated trace file is closed and the trace is resumed with output
to the new file specified with <tracefile>. Contact GoldenGate
Technical Support for assistance if the trace reveals significant 
processing bottlenecks. Turn off tracing with TRACE OFF or TRACE2 OFF.

TRACEINIT
Initializes (resets) tracing statistics back to 0 and then starts
accumulating statistics again. Use this option to track the current
behavior of processing, as opposed to historical.

TRANLOGOPTIONS PURGEORPHANEDTRANSACTIONS | 
NOPURGEORPHANEDTRANSACTIONS
Valid for Oracle RAC. Enables or disables purging of orphaned
transactions that occur when a node fails and Extract cannot capture
the rollback. Also can be controlled by the TRANLOGOPTIONS parameter
in Extract.

TRANLOGOPTIONS TRANSCLEANUPFREQUENCY <minutes>
Valid for Oracle RAC. Specifies the interval, in minutes, after which
GoldenGate scans for orphaned transactions and then re-scans to confirm
and delete them. Valid values are from 1 to 43200 minutes. Default is
10 minutes. 


VAMMESSAGE <Teradata command> |
VAMMESSAGE {ARSTATS | INCLUDELIST [filter] |
   EXCLUDELIST [filter]} |
VAMMESSAGE OPENTRANS

Sends a command to a Vendor Access Module (VAM).

<Teradata <command> can be:

control:terminate
Stops a replication group. Required before dropping or altering a
replication group in Teradata.

control:suspend
Suspends a replication group. Can be used when upgrading GoldenGate.

control:resume
Resumes a replication group after it has been suspended.

control:copy <database>.<table>
Copies a table from the source database to the target database.

SQL/MX commands can be:

ARSTATS
Displays TMF audit reading statistics

INCLUDELIST [filter] 
Displays the list of tables for which Extract has encountered data
records in the audit trail that match the selection criteria in the
TABLE parameters. [filter] is an optional wildcard pattern to filter
the list of tables returned.

EXCLUDELIST [filter] 
Displays the list of tables for which Extract has encountered data
records in the audit trail that do not match the selection criteria in
the TABLE parameters. [filter] is an optional wildcard pattern to
filter the list of tables returned. Certain system tables that are
implicitly excluded will always be present in the list of excluded
tables.
The module returns a response to GGSCI. The response can be either
ERROR or OK along with a response message.

SQL Server command can be:

OPENTRANS
Prints a list of open transactions with their transaction ID, start
time, first LSN, and the number of operations they contain.


Example 1: SEND EXTRACT finance, ROLLOVER
Example 2: SEND EXTRACT finance, STOP
Example 3: SEND EXTRACT finance, VAMMESSAGE control:suspend
Example 4: SEND EXTRACT finance, SHOWTRANS COUNT 10
Example 5: SEND EXTRACT finance, SKIPTRANS 5.17.27634 THREAD 2

----------------------------------------------------------------------
START EXTRACT

Use START EXTRACT to start the Extract process. To confirm that 
Extract has started, use the INFO EXTRACT or STATUS EXTRACT command.
Extract also can be started from the operating systems command line
for certain synchronization configurations. For more information on
the proper configuration and startup method to use for your purposes,
see the GoldenGate for Windows and UNIX Operations Guide.

Syntax:	
START EXTRACT <group name>

<group name>
The name of an Extract group or a wildcard (*) to specify multiple 
groups. For example, T* starts all Extract groups whose names
start with T.

Example: START EXTRACT finance

---------------------------------------------------------------------
STATS EXTRACT

Use STATS EXTRACT to display Extract statistics. To get the most
accurate number of operations per second that are being processed,
issue the STATS EXTRACT command with the RESET option, then wait a
few minutes and issue the STATS EXTRACT REPORTRATE command. The
LATEST STATISTICS field shows the operations per second.
 
Syntax:
STATS EXTRACT <group name>
[, <statistic>]
[, TABLE <table>]
[, TOTALSONLY <table spec>]
[, REPORTFETCH | NOREPORTFETCH]
[, REPORTRATE <time units>]
[, ...]

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* returns statistics for all Extract groups
whose names start with T.

<statistic>
The statistic to be displayed. Valid values are:
   TOTAL: Displays totals since processing startup.
   DAILY: Displays totals since the start of the current day.
   HOURLY: Displays totals since the start of the current hour.
   LATEST: Displays totals since the last RESET command.
   RESET: Resets the counters in the LATEST statistical field.
More than one statistic can be specified by separating each with a
comma, for example STATS EXTRACT finance, TOTAL, DAILY.
	
TABLE <table>
Displays statistics only for the specified table or a group of tables
specified with a wildcard (*).
	
TOTALSONLY <table spec>
Totals statistics for the specified table or a group of tables 
specified with a wildcard (*).
	
REPORTFETCH | NOREPORTFETCH
Controls whether or not statistics about fetch operations are included
in the output. The default is NOREPORTFETCH.

REPORTRATE <time  units>
Displays statistics in terms of processing rate rather than absolute 
values. Valid values are HR, MIN, or SEC.

Example: The following example displays total and hourly statistics 
per minute for a table and also resets the latest statistics.

STATS EXTRACT finance, TOTAL, HOURLY, TABLE acct, REPORTRATE MIN, 
RESET
			
---------------------------------------------------------------------
STATUS EXTRACT

Use STATUS EXTRACT to determine whether or not Extract is running.

Syntax:	
STATUS EXTRACT <group name> 
[, TASKS] 
[, ALLPROCESSES]

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* returns status for all Extract groups 
whose names start with T.

TASKS
Displays status only for Extract tasks. By default, tasks are not 
displayed unless you specify a single Extract group (without
wildcards).

ALLPROCESSES
Displays status for all Extract groups, including tasks.

Example 1: STATUS EXTRACT finance
Example 2: STATUS EXTRACT fin*

---------------------------------------------------------------------
STOP EXTRACT

Use STOP EXTRACT to stop Extract gracefully. The command 
preserves the state of synchronization for the next time Extract
starts, and it ensures that Manager does not automatically start
Extract.

Syntax:
STOP EXTRACT <group name>

<group name>
The name of an Extract group or a wildcard (*) to specify multiple
groups. For example, T* stops all Extract groups whose names
start with T.


Example: STOP EXTRACT finance

---------------------------------------------------------------------
####################################
#
#TRAIL COMMANDS
#
#
####################################
---------------------------------------------------------------------
ADD EXTTRAIL

Use ADD EXTTRAIL to create a trail for online processing on the local 
system and:
* Associate it with an Extract group.
* Assign a maximum file size.

Note: If the target database is DB2 running on z/OS and you
Will be using a local trail with a Replicat process that reads from
This trail on the same machine, do not use ADD EXTTRAIL. Instead, use
The ADD RMTTRAIL command. ADD EXTTRAIL is not compatible with
Converting data to EBCDIC because it bypasses the Collector process,
Which performs the conversion.

Syntax:
ADD EXTTRAIL <trail name>, EXTRACT <group name> [,MEGABYTES <n>]

<trail name>
The fully qualified path name of the trail. The actual trail name
can contain only two characters. GoldenGate appends this name with
a six-digit sequence number whenever a new file is created. For
example, a trail named /ggs/dirdat/tr would have files named
/ggs/dirdat/tr000001, /ggs/dirdat/tr000002, and so forth. 

<group name>
The name of the Extract group to which the trail is bound. Only
one Extract process can write data to a trail.

MEGABYTES <n>
The maximum size, in megabytes, of a file in the trail. The
default is 10. 

Example: ADD EXTTRAIL c:\ggs\dirdat\aa, EXTRACT finance, MEGABYTES 20

---------------------------------------------------------------------
ADD RMTTRAIL

Use ADD RMTTRAIL to create a trail for online processing on a remote
system and:
* Associate it with an Extract group.
* Assign a maximum file size.

In the parameter file, specify a RMTHOST entry before any RMTTRAIL
entries to identify the remote system and TCP/IP port for the Manager
process.

Syntax:
ADD RMTTRAIL <trail name>, EXTRACT <group name> [, MEGABYTES <n>]

<trail name>
The fully qualified path name of the trail. The actual trail name can
contain only two characters. GoldenGate appends this name with a six-
digit sequence number whenever a new file is created. For example, a
trail named /ggs/dirdat/tr would have files named /ggs/dirdat/tr000001,
/ggs/dirdat/tr000002, and so forth.

<group name>
The name of the Extract group to which the trail is bound. Only one
Extract process can write data to a trail.

MEGABYTES <n>
The maximum size, in megabytes, of a file in the trail. The default 
is 10. 

Example: ADD RMTTRAIL c:\ggs\dirdat\aa, EXTRACT finance, MEGABYTES 20

---------------------------------------------------------------------
ALTER EXTTRAIL

Use ALTER EXTTRAIL to change attributes of an existing trail that was
created with the ADD EXTTRAIL command (a trail on the local system).
The change takes effect the next time that Extract starts.

Syntax:
ALTER EXTTRAIL <trail name>, EXTRACT < group name> [, MEGABYTES <n>]

<trail name>
The fully qualified path name of the trail, for example 
c:\ggs\dirdat\aa.

<group name>
The name of the Extract group to which the trail is bound.

MEGABYTES <n>
The maximum size of a file, in megabytes. The default is 10. 

Example: ALTER EXTTRAIL c:\ggs\dirdat\aa, EXTRACT finance, MEGABYTES 20

-----------------------------------------------------------------------
ALTER RMTTRAIL

Use ALTER RMTTRAIL to change attributes for an existing trail that
was created with the ADD RMTTRAIL command (a trail on a remote
system). The change takes effect the next time that Extract starts.

Syntax:
ALTER RMTTRAIL <trail name>, EXTRACT <group name> [, MEGABYTES <n>]

<trail name>
The fully qualified path name of the trail, for example
c:\ggs\dirdat\aa.

<group name>
The name of the Extract group to which the trail is bound.

MEGABYTES <n>
The maximum size of a file, in megabytes. The default is 10.

Example: 
ALTER RMTTRAIL c:\ggs\dirdat\aa, EXTRACT finance, MEGABYTES 20

---------------------------------------------------------------------
DELETE EXTTRAIL

Use DELETE RMTTRAIL to delete the record of checkpoints associated with
a trail on the local system. Checkpoints are maintained in a file
bearing the same name as the group in the dirchk sub-directory of the
GoldenGate directory. This command only deletes references to the 
Specified trail from the checkpoint file. It does not delete the trail
Files themselves. To delete the trail files, use standard operating 
System commands for removing files.

Syntax:
DELETE EXTTRAIL <trail name>

<trail name>
The fully qualified path name of the trail, including the two-
character trail prefix.

Example:
DELETE EXTTRAIL c:\ggs\dirdat\aa

---------------------------------------------------------------------
DELETE RMTTRAIL

Use DELETE RMTTRAIL to delete the record of checkpoints associated with
a trail on a remote system. Checkpoints are maintained in a file
bearing the same name as the group in the dirchk sub-directory of the
GoldenGate directory. This command only deletes references to the 
Specified trail from the checkpoint file. It does not delete the trail
Files themselves. To delete the trail files, use standard operating 
System commands for removing files.

Syntax:
DELETE RMTTRAIL <trail name>

<trail name>
The fully qualified path name of the trail. 

Example: DELETE RMTTRAIL c:\ggs\dirdat\aa

---------------------------------------------------------------------
INFO EXTTRAIL

Use INFO EXTTRAIL to retrieve configuration information for a local
trail. It shows the name of the trail, the Extract that writes to it,
the position of the last data processed, and the assigned maximum file
size.

Syntax:
INFO EXTTRAIL <trail name>

<trail name>
The fully qualified path name of the trail or a wildcard designating
multiple trails. 

Example 1:	INFO EXTTRAIL c:\ggs\dirdat\aa
Example 2:	INFO EXTTRAIL *

---------------------------------------------------------------------
INFO RMTTRAIL

Use INFO RMTTRAIL to retrieve configuration information for a remote
trail. It shows the name of the trail, the Extract that writes to it, 
the position of the last data processed, and the assigned maximum file 
size.

Syntax:
INFO RMTTRAIL <trail name>

<trail name>
The fully qualified path name of the trail or a wildcard designating
multiple trails. 

Example 1:	INFO RMTTRAIL c:\ggs\dirdat\aa
Example 2:	INFO RMTTRAIL *

---------------------------------------------------------------------
####################################
#
#MANAGER COMMANDS
#
#
####################################
---------------------------------------------------------------------
INFO MANAGER

Use INFO MANAGER to determine whether or not the Manager process is
running. If Manager is running, the port number is displayed. This
command is an alias for STATUS MANAGER.

Syntax:
INFO MANAGER

---------------------------------------------------------------------
REFRESH MANAGER

Use REFRESH MANAGER to execute the Manager parameter file. This
command enables you to make parameter changes to the Manager
configuration without stopping and restarting the Manager process.
The exception is changing the port number. To change the Manager port
number, you must first stop Manager with the STOP MANAGER command. 
Edit the Manager parameter file to change the port number and any 
other parameters as needed, and then start Manager again with the 
START MANAGER command.

Syntax:
REFRESH MANAGER

---------------------------------------------------------------------
SEND MANAGER

Use SEND MANAGER to retrieve the status of the active Manager
process, or to retrieve dynamic port information as configured in the
Manager parameter file.

Syntax:
SEND MANAGER
[CHILDSTATUS [DEBUG]]
[GETPORTINFO [DETAIL]]
[GETPURGEOLDEXTRACTS]

CHILDSTATUS [DEBUG]
Retrieves status information about processes started by Manager. 
DEBUG returns the port numbers that have been allocated by Manager.
(Requires ports to be specified with the DYNAMICPORTLIST parameter.)
 
GETPORTINFO [DETAIL]
Retrieves the current list of ports that have been allocated by
Manager and their corresponding process IDs. (Requires ports to be
specified with the DYNAMICPORTLIST parameter.)

GETPURGEOLDEXTRACTS
Displays current trail-maintenance rules, as set with the
PURGEOLDEXTRACTS parameter in the Manager parameter file.

Example 1:	SEND MANAGER CHILDSTATUS
Example 2:	SEND MANAGER GETPORTINFO

---------------------------------------------------------------------
START MANAGER

Use START MANAGER to start the Manager process.

Syntax:
START MANAGER

---------------------------------------------------------------------
STATUS MANAGER

Use STATUS MANAGER to determine whether or not the Manager process is
running. If Manager is running, the port number is displayed.

Syntax:
STATUS MANAGER

---------------------------------------------------------------------
STOP MANAGER

Use STOP MANAGER to stop the Manager process. Stopping Manager stops
all other GoldenGate processes, which affects any synchronization
activities in process.
 
Syntax:
STOP MANAGER [!]

!
(Exclamation point) Bypasses the prompt confirming the intent to shut
down Manager.

---------------------------------------------------------------------
####################################
#
#PARAM COMMANDS
#
#
####################################

---------------------------------------------------------------------
EDIT PARAMS

Use EDIT PARAMS to create or change a parameter file. By default,
this command launches Microsoft Notepad on Windows or the vi editor
on UNIX systems. You can change the editor with the SET EDITOR
command.

Syntax:
EDIT PARAMS {MGR | <group> | <file name>}

MGR
Opens a parameter file for the Manager process.

<group>
Opens a parameter file for the specified Extract or Replicat group.

<file name>
Opens the specified file. Use the full path name.

Example 1:	EDIT PARAMS finance
Example 2:	EDIT PARAMS c:\lpparms\replp.prm

---------------------------------------------------------------------
SET EDITOR

Use SET EDITOR to change the default text editor for the current
session of GGSCI. The default editors are Microsoft Notepad on Windows
and the Vi editor on UNIX.

Syntax:
SET EDITOR <program name>

<program name>
Any editor on the system that can save text in ASCII format.

Example: The following example changes the default editor to
Wordpad.

SET EDITOR wordpad

---------------------------------------------------------------------
VIEW PARAMS

Use VIEW PARAMS to view the contents of a parameter file.

Syntax:
VIEW PARAMS {MGR | <group> | <file name>}

MGR
Shows the Manager parameter file.

<group>
Shows the parameter file for the specified Extract or Replicat group.

<file name>
Shows the specified file. Use the full path name.

Example 1:	VIEW PARAMS finance
Example 2:	VIEW PARAMS c:\lpparms\replp.prm

---------------------------------------------------------------------
####################################
#
#REPLICAT COMMANDS
#
#
####################################


---------------------------------------------------------------------
ADD REPLICAT

Use ADD REPLICAT to create a Replicat group. Unless a special run is
specified, ADD REPLICAT creates checkpoints so that processing
continuity is maintained from run to run. See the GoldenGate 
Operations Guide for Windows and UNIX for procedures that include 
creating Replicat groups.

Syntax:
ADD REPLICAT <group name>
{, SPECIALRUN | 
   , EXTFILE <full path name> | 
   , EXTTRAIL <full path name>} 
[, BEGIN <start point> |
   , EXTSEQNO <seqno>, EXTRBA <relative byte address>]
[, CHECKPOINTTABLE <owner.table> | NODBCHECKPOINT]
[, PARAMS <parameter file>]
[, REPORT <report file>]
[, DESC <description>]

<group name>
The name of the Replicat group. Follow these naming conventions:
* You can use up to eight ASCII characters, including nonalphanumeric
  characters such as the underscore (_). Any ASCII character can be
  used, so long as the operating system allows that character to be
  in a filename. This is because a group is identified by its
  associated checkpoint file.
* Group names are not case-sensitive.
* Use only one word.
* Do not use the word port as a group name. However, you can use
  the string port as part of the group name.
* Do not place a numeric value at the end of a group name, such as
  fin1, fin10, and so forth. You can place a numeric value at the
  beginning of a group name, such as 1_fin, 1fin, and so forth.
   ext_1
   ex+2t
   ex!2t

SPECIALRUN
Creates a Replicat special run. Either SPECIALRUN, EXTFILE, 
or EXTTRAIL is required.

EXTFILE <full path name>
Specifies an extract file that was created with the ADD RMTFILE 
command.

EXTTRAIL <full path name>
Specifies a trail that was created with the ADD RMTTRAIL or ADD
EXTTRAIL command.

BEGIN <start point>
Defines an initial checkpoint in the trail. 
*  To begin replicating changes from when the ADD REPLICAT 
   command was issued, use the NOW argument. 
*  To begin extracting changes from a specific time, use the 
   date-time format of YYYY-MM-DD HH:MM[:SS[.CCCCCC]].

EXTSEQNO <seqno>
Specifies the sequence number of the file in a trail in which to begin
processing data. Specify the sequence number, not any zeroes used for
padding. For example, if the trail file is c:\ggs\dirdat\aa000026, you
would specify EXTSEQNO 26. By default, processing begins at the
beginning of a trail unless this option is used. To use EXTSEQNO, you
must also use EXTRBA. Contact GoldenGate Technical Support before using
this option.

EXTRBA <relative byte address>
Specifies the relative byte address within the trail file specified by
EXTSEQNO. Contact GoldenGate Technical Support before using this
option.

CHECKPOINTTABLE <owner.table>
Specifies that this Replicat group will write checkpoints to the
specified table in the database. Include the owner and table name, as
in hr.hr_checkpoint. This argument overrides the default
CHECKPOINTTABLE specification in the GLOBALS file. The table must be
added with the ADD CHECKPOINTTABLE command.

NODBCHECKPOINT
Specifies that this Replicat group will not write checkpoints to a
checkpoint table. This argument overrides the default CHECKPOINTTABLE
specification in the GLOBALS file.

PARAMS <parameter file>
Specifies a parameter file in a location other than the default of
dirprm within the GoldenGate directory. Use the fully qualified path
name.

REPORT <report file>
Specifies a Replicat report file in a location other than the default
of dirrpt within the GoldenGate directory. Use the fully qualified
path name.

DESC <description>
Specifies a description of the group, such as "Loads ACCOUNT_TAB on
Serv2". The description must be within quotes.

Example: ADD REPLICAT sales, EXTTRAIL d:\ggs\dirdat\rt

---------------------------------------------------------------------
ALTER REPLICAT

Use ALTER REPLICAT to change the attributes of a Replicat group that
was created by the ADD REPLICAT command. The CHECKPOINT and
NODBCHECKPOINT options cannot be changed with this command. Before
using this command, stop Replicat by issuing the STOP REPLICAT <group
name> command.

Syntax:
ALTER REPLICAT <group name> [, <option>] [, ...] 

<group name>
The name of a Replicat group.
 
[<option>] can be any of the following. If no options are specified, 
the checkpoint is reset to the beginning of the trail. 

EXTSEQNO <seqno>
Specifies the sequence number of the file in a trail in which to begin
processing data. Specify the sequence number, not any zeroes used for
padding. For example, if the trail file is c:\ggs\dirdat\aa000026, you
would specify EXTSEQNO 26. By default, processing begins at the
beginning of a trail unless this option is used. To use EXTSEQNO, you
must also use EXTRBA. Contact GoldenGate Technical Support before using
this option.

EXTRBA <relative byte address>
Specifies the relative byte address within the trail file specified by
EXTSEQNO. Contact GoldenGate Technical Support before using this
option.

PARAMS <parameter file>
Specifies a parameter file in a location other than the default of
dirprm within the GoldenGate directory. Use the fully qualified path
name of the file.

REPORT <report file>
Specifies a Replicat report file in a location other than the default
of dirrpt within the GoldenGate directory. Use the fully qualified
path name of the file.

DESC <description>
Specifies a description of the group, such as "Loads ACCOUNT_TAB on
Serv2". The description must be within quotes.

Example 1:	ALTER REPLICAT, finance EXTSEQNO 53
Example 2:	ALTER REPLICAT, finance EXTRBA 0

---------------------------------------------------------------------
CLEANUP REPLICAT

Use CLEANUP REPLICAT to delete run history for a specified Replicat 
group. The cleanup keeps the last run record intact so that Replicat
can resume processing from where it left off. Before using this
command, stop Replicat by issuing the STOP REPLICAT <group name>
command.

Syntax:
CLEANUP REPLICAT <group name> [, SAVE <count>]

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* cleans up all Replicat groups whose names
start with T.

SAVE <count>
Excludes the specified number of the most recent records from the 
cleanup.

Example 1: The following deletes all but the last record.
CLEANUP REPLICAT finance

Example 2:The following deletes all but the most recent five records
CLEANUP REPLICAT *, SAVE 5

---------------------------------------------------------------------
DELETE REPLICAT

Use DELETE REPLICAT to delete a Replicat group. This command has the
side effect of freeing up trail files for purging by Manager, because
the checkpoints used by the deleted group are removed (assuming no
other processes are reading the file). Before using this command,
stop Replicat by issuing the STOP REPLICAT <group name> command.

Before using this command, do the following:
1. Stop Replicat.
     STOP REPLICAT <group name> 
2. If this group is using a database checkpoint table, log into the
   database by using the DBLOGIN command, so that the checkpoints
   can be deleted from the table.

Syntax:
DELETE REPLICAT <group name>  [!]

<group name>
The name of a Replicat group or a set of group names specified with
a wildcard (*). For example, T* deletes all Replicat groups starting
with T.

!
Use this option to delete this groups checkpoints from the checkpoint
file on disk, but not from the checkpoint table in the database. If
there is an open Replicat transaction, allowing database checkpoints to
be deleted causes the transaction to be committed, and this option can
be used to prevent that.

Example: DELETE REPLICAT finance

---------------------------------------------------------------------
INFO REPLICAT

Use INFO REPLICAT to retrieve processing history for a Replicat
group. The output of this command includes:

* The status of Replicat (STARTING, RUNNING, STOPPED or ABENDED).
* Approximate replication lag.
* The trail from which Replicat is reading.
* Replicat run history, including checkpoints in the trail.
* Information about the Replicat environment.

Byte lag is the difference, in bytes, between the read position of
Replicat in the trail file at the time of the last checkpoint and the 
current end-of-file indicator.
 
Time lag is the lag, in seconds, at the time the last checkpoint was
written to the trail. For example, if the current time is 15:00:00, 
and the last checkpoint was at 14:59:00, and the timestamp of the 
last record processed was 14:58:00, then the lag is 00:01:00 (one
minute - the difference between 14:58 and 14:59).

A lag value of UNKNOWN indicates that Replicat might have
started but has not yet processed records, or that the source 
system's clock is ahead of the target system's clock (due to clock
imperfections, not time zone differences).

For more precise lag information, use LAG REPLICAT.

The basic INFO REPLICAT command, without the TASKS or ALLPROCESSES 
argument, displays information only for online (continuous) Replicat
groups. Tasks are excluded.

Syntax:
INFO REPLICAT <group name>
[, DETAIL]
[, SHOWCH [<n>]]
[, TASKS | ALLPROCESSES]
[, ...]

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* shows information for all Replicat groups
whose names start with T.
 
DETAIL
Displays detail information.

SHOWCH [<n>]
Displays current checkpoint details, including those recorded to the
checkpoint file and those recorded to the checkpoint table, if one is
being used. The database checkpoint display includes the table name,
the hash key (unique identifier), and the create timestamp. Specify a
value for <n> to include the specified number of previous checkpoints
as well as the current one.

TASKS
Displays only Replicat tasks. Tasks that were specified by a wildcard
argument are not displayed by INFO REPLICAT.

ALLPROCESSES
Display all Replicat groups, both online and tasks.

Example 1: INFO REPLICAT *, DETAIL, ALLPROCESSES
Example 2: INFO REPLICAT *, TASKS
Example 3: INFO REPLICAT finance
Example 4: INFO REPLICAT finance, SHOWCH

---------------------------------------------------------------------
KILL REPLICAT

Use KILL REPLICAT to kill a Replicat process. Killing a process 
leaves the most recent checkpoint in place, and the current transaction
is rolled back by the database, guaranteeing that no data is lost when
the process is restarted. The Manager process will not attempt to
restart a killed Replicat process. Use this command only if Replicat
cannot be stopped gracefully with the STOP REPLICAT command.

Syntax:
KILL REPLICAT <group name>

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* kills all Replicat groups whose names
start with T.

Example: KILL REPLICAT finance

---------------------------------------------------------------------
LAG REPLICAT

Use LAG REPLICAT to determine a true lag time between Replicat
processing and the trail. Lag time is the difference, in seconds,
between the time that the last record was processed by Replicat 
and the timestamp of the record in the trail.

LAG REPLICAT estimates lag time more precisely than INFO REPLICAT
because it communicates with the Replicat process directly rather
than reading a checkpoint position.

Syntax:
LAG REPLICAT <group name>

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* shows lag for all Replicat groups whose
names start with T.

Example 1:	LAG REPLICAT *
Example 2:	LAG REPLICAT *fin*

---------------------------------------------------------------------
SEND REPLICAT

Use SEND REPLICAT to communicate with a running Replicat process. The
request is processed as soon as the group is ready to accept commands
from users.

Syntax:
SEND REPLICAT <group name>, 
FORCESTOP |
GETLAG |
HANDLECOLLISIONS [<table spec>] |
NOHANDLECOLLISIONS [<table spec>] |
REPORT [HANDLECOLLISIONS [<table spec>]] |
STATUS |
STOP |
TRACE[2] [DDLINCLUDE | DDL[ONLY]] [FILE] <tracefile> |
TRACE[2] OFF |
TRACEINIT

<group name>
The name of the Replicat group. If the group is not running, an error
is returned.

FORCESTOP
Forces Replicat to stop, bypassing any notifications. This command will
roll back any active transaction and stop the process immediately.

GETLAG
Shows a true lag time between Replicat and the trail. Lag time is the
difference, in seconds, between the time that the last record was
processed by Replicat and the timestamp of the record in the  trail.
The results are the same as LAG EXTRACT.

HANDLECOLLISIONS [<table spec>]
Turns on the HANDLECOLLISIONS parameter. Instead of using this option,
you can specify the HANDLECOLLISIONS parameter in the Replicat
parameter file. HANDLECOLLISIONS is used for automatic error handling
when performing initial data loads while the source database is active.
Make certain to disable HANDLECOLLISIONS (either with SEND REPLICAT or
by removing the parameter from the parameter file) after the initial
load is complete and online data changes have been applied to the
target tables. <table spec> restricts HANDLECOLLISIONS to a specific
target table or a group of target tables specified with a standard
wildcard (*).

NOHANDLECOLLISIONS [<table spec>]
Turns off the HANDLECOLLISIONS parameter but does not remove it from
the parameter file. To avoid enabling HANDLECOLLISIONS the next time
Replicat starts, remove it from the parameter file.
<table spec> restricts NOHANDLECOLLISIONS to a specific target table or
a group of target tables specified with a standard wildcard (*).  

REPORT [HANDLECOLLISIONS [<table spec>]]
Generates an interim statistical report to the Extract report file. 
The statistics that are displayed depend upon the configuration of the 
STATOPTIONS parameter when used with the RESETREPORTSTATS |
NORESETREPORTSTATS option.
HANDLECOLLISIONS shows tables for which
HANDLECOLLISIONS has been enabled. <table spec> restricts the output to
a specific target table or a group of target tables specified with a
standard wildcard (*).

STATUS
Returns the current location within the trail and information
about the current transaction.

STOP
Stops Replicat.

TRACE[2] [DDLINCLUDE | DDL[ONLY]] [FILE] <tracefile> 
Turns tracing on and off. Tracing captures information to the
specified file to reveal processing bottlenecks. TRACE captures 
step-by-step processing information. TRACE2 identifies code segments
rather than specific steps. If a trace already is in progress, the
associated trace file is closed and the trace resumes to the file
specified with <tracefile>. Contact GoldenGate Technical Support for
assistance if the trace reveals significant processing bottlenecks.
Turn off tracing with TRACE OFF or TRACE2 OFF.

  * DDLINCLUDE | DDLONLY (Replicat only) enables DDL tracing and
    specifies how DDL tracing is included in the trace report.
    DDLINCLUDE includes DDL tracing in addition to regular tracing of
    transactional data processing. DDL[ONLY] excludes the tracing of
    transactional data processing and only traces DDL. This option can
    be abbreviated to DDL.

  * [FILE] <file name> specifies the fully qualified name of a file to
    which GoldenGate logs the trace information. The FILE keyword is
    optional, but must be used if other parameter options will follow
    the file name, for example:
    SEND REPLICAT <group> TRACE FILE <file name> DDLINCLUDE
    If no other options will follow the file name, the FILE keyword can
    be omitted, for example:
    SEND REPLICAT <group> TRACE DDLINCLUDE <file name>

TRACEINIT
Initializes (resets) tracing statistics back to 0 and then starts
accumulating statistics again. Use this option to track the current
behavior of processing, as opposed to historical.

Example 1: SEND REPLICAT finance, REPORT
Example 2: SEND REPLICAT finance, HANDLECOLLISIONS
Example 3: SEND REPLICAT finance, REPORT HANDLECOLLISIONS fin_*
Example 4: SEND REPLICAT finance, GETLAG

---------------------------------------------------------------------
START REPLICAT

Use START REPLICAT to start Replicat. To confirm that Replicat has
started, use the INFO REPLICAT or STATUS REPLICAT command.

START REPLICAT, without any options, causes Replicat to start
processing at one of the following points to maintain data integrity:

*After graceful or abnormal termination: At the last unprocessed
 transaction in the trail from the previous run, as represented by the
 current read checkpoint.

* First-time startup after the group was created: From the
  beginning of the active trail file (seqno 0, rba 0).

The SKIPTRANSACTION, ATCSN, and AFTERCSN options of START REPLICAT
cause Replicat to begin processing at a transaction in the trail other
than the normal start point.

Replicat also can be started from the operating system's command line
for certain synchronization configurations. For more information on 
the proper configuration and startup method to use for your purposes,
see the GoldenGate for Windows and UNIX Operations Guide.

Syntax:
START REPLICAT <group name> 
[SKIPTRANSACTION | ATCSN <csn> | AFTERCSN <csn>]

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* starts all Replicat groups whose names 
start with T.

SKIPTRANSACTION
Causes Replicat to skip the first transaction after its expected
startup position in the trail. All operations from that first
transaction are excluded. 
If the MAXTRANSOPS parameter is also being used for this Replicat, it
is possible that the process will start to read the trail file from
somewhere in the middle of a transaction. In that case, the remainder
of the partial transaction is skipped, and Replicat resumes normal
processing from the next begin-transaction record in the file. The
skipped records are written to the discard file if the DISCARDFILE
parameter is being used; otherwise, a message is written to the report
file.

Limitations:
* Not valid when SPECIALRUN is used in ADD REPLICAT or in the Replicat
  parameter file.
* Not valid if this Replicat is the object of an Extract that is
  configured with the RMTTASK parameter.

ATCSN <csn>
Causes Replicat to skip transactions in the trail until it finds a
begin-transaction indicator that contains the specified commit sequence
number (CSN). This transaction and subsequent ones are applied to the
target. All transactions with a CSN less than the specified one are
skipped.

For <csn>, supply the CSN that is native to the database.
CSN is a generic GoldenGate term that means the sequence number that is
used by the database to identify transactions.

CSN formats by database:

DB2 LUW: <time>.<LSN>
DB2 z/OS: <RBA>
Oracle: <system change number>
SQL Server: <LSN>:<segment offset>:<slot> 
Ingres: <LSN-high>:<LSN-low>
Sybase: <time_high>.<time_low>.<page>.<row>
c-tree: <log number>|<byte offset>
SQL/MX: <sequence number>|<RBA>
Teradata:


When this option is used, a message is written to the report file.

Limitations:
* Not valid when SPECIALRUN is used in ADD REPLICAT or in the Replicat
  parameter file.
* Not valid if this Replicat is the object of an Extract that is
  configured with the RMTTASK parameter.
* To support starting at a CSN, the trail must be of GoldenGate version
  10.0 or later, because the CSN is stored in the file header. If
  Replicat is started with ATCSN against an earlier trail version,
  Replicat will abend and write an error to the report stating that the
  trail format is not supported.


AFTERCSN <csn>
Causes Replicat to skip transactions in the trail until it finds the
first transaction after the one that contains the specified CSN. All
transactions whose begin-transaction record in the trail contain a CSN
less than, or equal to, the specified one are skipped. 

For <csn>, supply the CSN that is native to the database.
CSN is a generic GoldenGate term that means the sequence number that is
used by the database to identify transactions.

CSN formats by database:

DB2 LUW: <time>.<LSN>
DB2 z/OS: <RBA>
Oracle: <system change number>
SQL Server: <LSN>:<segment offset>:<slot> 
Ingres: <LSN-high>:<LSN-low>
Sybase: <time_high>.<time_low>.<page>.<row>
c-tree: <log number>|<byte offset>
SQL/MX: <sequence number>|<RBA>
Teradata:

When this option is used, a message is written to the report file.

Limitations:
* Not valid when SPECIALRUN is used in ADD REPLICAT or in the Replicat
  parameter file.
* Not valid if this Replicat is the object of an Extract that is
  configured with the RMTTASK parameter.
* To support starting after a CSN, the trail must be of 
  GoldenGate version 10.0 or later, because the CSN is stored in the
  file header. If Replicat is started with AFTERCSN against an earlier
  trail version, Replicat will abend and write an error to the 
  report stating that the trail format is not supported.

NOTE: When a record that is specified with a CSN is found, Replicat
issues a checkpoint to ensure that subsequent restarts of the process
that occur before the next checkpoint will start from the requested
location, and not from a point prior to the requested CSN.

For more information about the CSN, see the GoldenGate for Windows and
UNIX Administrator Guide.

Example 1: START REPLICAT finance

Example 2: The following starts Replicat at an Oracle-specific CSN.
START REPLICAT finance, ATCSN 6488359

Example 3: The following starts Replicat at a SQL Server-specific CSN.
START REPLICAT finance, AFTERCSN 000004D2:0000162E:0009


---------------------------------------------------------------------
STATS REPLICAT

Use STATS REPLICAT to display Replicat statistics.

Syntax:
STATS REPLICAT <group name>
[, <statistic>]
[, TABLE <table>]
[, TOTALSONLY <table>]
[, REPORTDETAIL | NOREPORTDETAIL]
[, REPORTRATE <time units>]
[, ... ]

<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* returns statistics for all Replicat groups
whose names start with T.

<statistic>
The statistic to be displayed. Valid values are:
  TOTAL: Displays totals since process startup.
  DAILY: Displays totals since the start of the current day.
  HOURLY: Displays totals since the start of the current hour.
  LATEST: Displays totals since the last RESET command.
  RESET: Resets the counters in the LATEST statistical field.
 
More than one statistic can be specified by separating each with a
comma, for example: STATS REPLICAT finance, TOTAL, DAILY.

TABLE <table>
Displays statistics only for the specified table or a group of tables
specified with a wildcard (*).

TOTALSONLY <table>
Totals statistics for the specified table or a group of tables
specified with a wildcard (*).

REPORTDETAIL | NOREPORTDETAIL
Controls whether or not the output includes operations that were not
replicated as the result of collision errors. These operations are
reported in the regular statistics (inserts, updates, and deletes
performed) plus as statistics in the detail display, if enabled. For
example, if 10 records were insert operations and they were all ignored
due to duplicate keys, the report would indicate that there were 10
inserts and also 10 discards due to collisions. The default is
REPORTDETAIL.

REPORTRATE <time units>
Displays statistics in terms of processing rate, rather than absolute
values. Valid values are HR, MIN, SEC.

Example: The following example displays total and hourly statistics
per minute for a specific table, and it also resets the latest
statistics.

STATS REPLICAT finance, TOTAL, HOURLY, TABLE acct, REPORTRATE MIN,
RESET

---------------------------------------------------------------------
STATUS REPLICAT

Use STATUS REPLICAT to determine whether or not Replicat is running.

Syntax:
STATUS REPLICAT <group name>
[, TASKS]
[, ALLPROCESSES]
 
<group name>
The name of a Replicat group or a wildcard (*) to specify multiple
groups. For example, T* returns status for all Replicat groups 
whose names start with T.

TASKS
Displays status only for Replicat tasks. By default, tasks are not
displayed unless you specify a single Replicat group (without
wildcards).

ALLPROCESSES
Displays status for all Replicat groups, including tasks.

Example 1: STATUS REPLICAT finance
Example 2: STATUS REPLICAT fin*

---------------------------------------------------------------------
STOP REPLICAT

Use STOP REPLICAT to stop Replicat gracefully. This command 
preserves the state of synchronization for the next time that 
Replicat starts, and it ensures that Manager does not automatically 
start Replicat.

Syntax:
STOP REPLICAT <group name> [!]

<group name>
The name of a Replicat group or a wildcard (*) specifying multiple
groups. For example, T* stops all Replicat groups whose names start
with T.

!
(Exclamation point) Stops Replicat immediately.

Example: STOP REPLICAT finance

---------------------------------------------------------------------
####################################
#
#TRACETABLE COMMANDS
#
#
####################################


---------------------------------------------------------------------
ADD TRACETABLE

Use ADD TRACETABLE to create a trace table in the Oracle database. 
The trace table prevents Replicat transactions from being extracted 
again in a bidirectional configuration. The default name of the trace
table is GGS_TRACE. 

Use the DBLOGIN command to establish a database connection before
using this command.

Syntax:
ADD TRACETABLE [<owner.table>]

<owner.table>
The owner and name of the trace table to be created. Use this option
only if the table name is different than the default of GGS_TRACE. The
owner must be the same owner that is specified with the USERID
parameter in the Extract parameter file. GoldenGate recommends using
the default table name.

When using a different name, it must be specified with the TRACETABLE 
parameter in the Extract and Replicat parameter files. Record the name 
because you will need it for the parameter files and, if necessary, to 
view statistics or delete the table.

Example 1: The following adds a trace table with the default name of 
GGS_TRACE.
 
ADD TRACETABLE

Example 2: The following adds a trace table with a user-defined name
of ora_trace.

ADD TRACETABLE ora_trace

---------------------------------------------------------------------
DELETE TRACETABLE

Use DELETE TRACETABLE to delete an existing trace table.  Use the
DBLOGIN command to establish a database connection before using this
command.

Syntax:
DELETE TRACETABLE [<owner.table>]

<owner.table>
The owner and name of the trace table to be deleted. An owner and name
are not required if the owner is the same as that specified with the
USERID parameter and the trace table has the default name of GGS_TRACE. 

Example: DELETE TRACETABLE ora_trace

---------------------------------------------------------------------
INFO TRACETABLE

Use the INFO TRACETABLE command to verify the existence of the
specified trace table in the local instance of the database. If 
the table exists, GoldenGate displays the name and the date and time
that it was created; otherwise GoldenGate displays a message that the
table does not exist.  Use the DBLOGIN command to establish a 
database connection before using this command.

Syntax:
INFO TRACETABLE [<owner.table>]

<owner.table>
The owner and name of the trace table to be verified. An owner and name
are not required if the owner is the same as that specified with the
USERID parameter and the trace table has the default name of GGS_TRACE.

Example: INFO TRACETABLE ora_trace

---------------------------------------------------------------------
####################################
#
#TRANDATA COMMANDS
#
#
####################################


---------------------------------------------------------------------
ADD TRANDATA

Use ADD TRANDATA  to enable GoldenGate to acquire the transaction
information it needs from the transaction logs. Use the DBLOGIN
command to establish a database connection before using this command

*** c-tree databases ***

You can use ADD TRANDATA to turn on the REPLICATE attribute of a c-tree
file without bouncing the c-tree server. As an alternative, you can
turn on REPLICATE by setting one or more REPLICATE parameters in the
c-tree server configuration file ctsrvr.cfg before starting the server.

*** DB2 databases ***

Use ADD TRANDATA to enable DATA CAPTURE CHANGES on specified tables.
This command supports DB2 LUW and DB2 z/OS. 

By default, ADD TRANDATA issues the following command to a DB2 z/OS
database:

ALTER TABLE < name> DATA CAPTURE CHANGES;


By default, ADD TRANDATA issues the following command to a DB2 LUW
database:

ALTER TABLE < name> DATA CAPTURE CHANGES INCLUDE LONGVAR COLUMNS;

For DB2 LUW, you can exclude LONGVAR columns by using ADD TRANDATA with
the EXCLUDELONG option. In that case, ADD TRANDATA issues the following
command:

ALTER TABLE < name> DATA CAPTURE CHANGES;

*** Sybase databases ***

ADD TRANDATA marks a Sybase table for replication by executing the
Sybase sp_setreptable system procedure. The default behavior for ADD
TRANDATA is to always propagate data if any column changes.

ADD TRANDATA options employ database features to control how the
database logs LOB data for the specified table. By default, Sybase
propagates all LOB information, and that information is
written to the trail whether or not it has changed.

*** SQL Server databases ***

Use ADD TRANDATA to provide the additional log information that
GoldenGate needs to reconstruct update operations. The SQL Server
transaction log does not provide enough information by default.


*** Oracle databases ***

Oracle 8.x

For Oracle 8.x, ADD TRANDATA installs an update trigger on the
specified table. When the trigger fires, it sets each key value to
itself, causing the values to be logged. The trigger is unobtrusive. It
only executes for update operations performed on the table where it is
installed, and it has virtually no effect on CPU load or database
performance. To use this trigger for tables in any schema other than
the GoldenGate schema, the GoldenGate user must be granted CREATE ANY
TRIGGER privileges.

Note: This trigger will not fire if applications update, insert, or 
delete LOBs by using the DBMS_LOB package or the Oracle OCI interface.

Oracle 9.x and later

For Oracle 9.x or later, ADD TRANDATA by default enables table-level 
supplemental logging, and the supplemental log group includes only the 
key columns. 

The command issues an ALTER TABLE command with an ADD SUPPLEMENTAL LOG
DATA clause that is appropriate for the type of unique constraint 
(or lack of one) that is defined for the table.

The following are additional options:

* Use the COLS option to log other columns as needed.
* Use the NOKEY option to prevent the logging of key columns when
  needed. When NOKEY is used, other columns that can serve as keys must
  be logged with COLS, and a KEYCOLS clause must be used in the TABLE
  or MAP statement to designate them as keys.
* To enforce the use of an update trigger instead of supplemental
  logging, use the USETRIGGER option. The use of triggers is required
  if the database compatibility is set to 8i.

Additional Oracle 9i or later requirements

One of the following additional steps must be taken when using ADD
TRANDATA for Oracle 9i or later, depending on the logging method that
was specified:

SUPPLEMENTAL LOGGING: Besides table-level logging, minimal supplemental 
logging must be enabled at the database level for GoldenGate to process 
updates to primary keys and chained rows. This is a known Oracle issue 
and must be done through the database interface, not through 
GoldenGate. To verify that supplemental logging is enabled at the 
database level, issue the following statement:

   SELECT SUPPLEMENTAL_LOG_DATA_MIN FROM V$DATABASE;
 
* For ORACLE 9i, the output of the query must be YES. When other
  LOG_DATA options are enabled, LOG_DATA_MIN is automatically enabled,
  but you should verify it to be certain. 
* For ORACLE 10g, the output of the query must be YES or IMPLICIT. As 
  of Oracle 10g, LOG_DATA_MIN must be explicitly set, because it is not 
  enabled automatically when other LOG_DATA options are set.

UPDATE TRIGGERS: If using ADD TRANDATA with the USETRIGGER option, add 
the TRANLOGOPTIONS parameter with the FETCHCHAINEDUPDATES option to the 
Extract parameter file.

Additional information for Oracle

* When you use an update trigger, any columns preceding the primary key
  columns are automatically included in the trigger definition and
  therefore are logged. This is normal behavior designed to
  accommodate row chaining.

* Oracle 8i has a limitation where, if a child table in a foreign-key
  relationship does not have an index on the foreign-key column, Oracle
  locks the table when the GoldenGate update trigger updates the key on
  that table. To prevent the locking, create an index on the foreign-
  key column.

* When creating an update trigger or supplemental log group with ADD 
  TRANDATA, GoldenGate forms a name as follows:

  Triggers: appends the table name, an underscore, and object ID to a 
  prefix of GGS_UB_. 
  Log groups: appends the table name, an underscore, and object ID to a 
  prefix of GGS_.

  Because Oracle limits an object name to 30 characters, GoldenGate 
  truncates long table names as needed so the prefix and object ID can 
  be included. For example, the name of a trigger on the 
  ACCT_HQ_A_SALES_COMP_TOTAL_DETAIL table would be 
  GGS_UB_ACCT_HQ_A_SALES_C_33276 .

  If you have upgraded from a version of GoldenGate prior to 8.0.2, see 
  the GoldenGate documentation for additional information about ADD 
  TRANDATA.

Syntax:

ADD TRANDATA <owner.table>
[, COLS (<column list>)]
[, INCLUDELONG | EXCLUDELONG]
[, LOBSNEVER | LOBSALWAYS | LOBSIFCHANGED] 
[, NOKEY]
[, OLDFORMAT]
[, USETRIGGER]

<owner.table>
The owner and name of the table or file for which the key is to be
logged. A wildcard can be used for the table name but not the owner
name.

INCLUDELONG | EXCLUDELONG
(DB2 LUW) Controls whether or not the ALTER TABLE issued
by ADD TRANDATA includes the "include longvar columns" attribute.
INCLUDELONG is the default. When ADD TRANDATA is issued with this
option, GoldenGate issues the following statement:

ALTER TABLE < name> DATA CAPTURE CHANGES INCLUDE LONGVAR COLUMNS;

When EXCLUDELONG is used, the following is the command:

ALTER TABLE < name> DATA CAPTURE CHANGES;

When EXCLUDELONG is used, GoldenGate does not support functionality
that requires before images of tables that include LONGVAR columns.
Examples of this functionality are GETUPDATEBEFORES, NOCOMPRESSUPDATES,
and NOCOMPRESSDELETES parameters. To support this functionality,
changes to LONGVAR columns in the DB2 transaction logs must include
both the before and after images of the column value. 

COLS (<column list>)
(Oracle) Adds additional, non-key column(s) to Oracles
supplemental logging so that the values appear in the transaction log.
Separate multiple columns with commas, for example NAME, ID, DOB.

LOBSNEVER | LOBSALWAYS | LOBSIFCHANGED
(Sybase) Control how the database propagates LOB data for the specified
table.
* LOBSNEVER prevents LOB data from being propagated.
* LOBSALWAYS always propagates LOB data whether or not it has changed
  in a transaction. This is the default.
* LOBSIFCHANGED only propagates LOB data if it was changed during a
  transaction.

NOKEY
(Oracle) Suppresses the logging of primary key columns. By 
default, the primary key is logged, if one exists. If this option is 
used, use the COLS option to log alternate columns, and designate those 
columns as keys by using the KEYCOLS option of the TABLE or MAP 
parameter.

OLDFORMAT
(Oracle) Uses the naming format for update triggers and log groups that 
Existed prior to GoldenGate version 8.0.2. This format appends GGS_ to 
Log group names and GGS_UB_ to update-trigger names created with ADD
TRANDATA, but does not include an object ID.

If the resultant name exceeds Oracles 30-character maximum, GoldenGate
truncates the table name. If more than one name gets truncated to the
same characters (when tables have similar names), there will be
duplicate triggers or log-groups, with no object ID to differentiate
them. This can cause INFO TRANDATA and DELETE TRANDATA to affect the
wrong object. Do not use this option without guidance from a GoldenGate
support representative.

USETRIGGER
(Oracle) Forces an update trigger to be used instead of Oracle's
supplemental logging (the default). Use this option for an Oracle 9i or 
later database if you prefer not to enable supplemental logging or if 
the database compatibility is set to 8i.

Example 1: The following example causes the primary key to be recorded
in the redo log for an Oracle table, logs additional data for a SQL 
Server table, marks a Sybase table for replication, or turns on the 
REPLICATE attribute of a c-tree file.

ADD TRANDATA finance.acct

Example 2: The following Oracle example causes the primary key to be
recorded plus the non-key columns NAME and ADDRESS.

ADD TRANDATA finance.acct, COLS (name, address)

Example 3: The following Oracle example prevents the primary key from
being logged, but logs the non-key columns NAME and PID instead.

ADD TRANDATA finance.acct, NOKEY, COLS (name, pid)

---------------------------------------------------------------------
DELETE TRANDATA

Use DELETE TRANDATA to do the following, depending on the database.

* c-tree: Disable replication for the specified file.

* DB2 LUW and DB2 on z/OS: Alters the table to DATA CAPTURE NONE.

* Oracle: Delete GoldenGate update triggers or disable supplemental
          logging.

* Sybase: Disable replication.

* SQL Server: Stop extended logging. 

Use the DBLOGIN command to establish a database connection before using
this command. The user specified with this command must have the same
privileges required for ADD TRANDATA.

Syntax:
DELETE TRANDATA <owner.table> [, OLDFORMAT] [, USETRIGGER]

<owner.table>
The owner and name of the table or file. A wildcard can be used for the
table name but not the owner name.

OLDFORMAT
Uses the naming format for update triggers and log groups that existed
prior to GoldenGate version 8.0.2. This format appends GGS_ to log
group names and GGS_UB_ to update-trigger names created with ADD
TRANDATA, but does not include an object ID.

USETRIGGER

Must be used to delete an update trigger if ADD TRANDATA <owner.table>,
USETRIGGER was used to add the trigger (Oracle 9i and later).

Example 1: DELETE TRANDATA finance.acct
Example 2: DELETE TRANDATA finance.ac*
Example 3: DELETE TRANDATA finance.acct, USETRIGGER

---------------------------------------------------------------------
INFO TRANDATA

Use INFO TRANDATA to get information about whether or not the 
effects of the ADD TRANDATA command are still active. The
effects of the ADD TRANDATA command differ per database.

Syntax:
INFO TRANDATA <owner.table> [, OLDFORMAT] [, USETRIGGER]

<owner.table>
The owner and name of the table or file for which you want to view
Trandata information.  The owner is not required if it is the same as
the user specified by the DBLOGIN command. A wildcard can be used for
the table name but not the owner name.

OLDFORMAT
Uses the naming format for update triggers and log groups that existed
prior to GoldenGate version 8.0.2. This format appends GGS_ to log
group names and GGS_UB_ to update-trigger names created with ADD
TRANDATA, but does not include an object ID.

USETRIGGER
Verifies whether or not a GoldenGate update trigger exists.

Example: INFO TRANDATA finance.acct


---------------------------------------------------------------------
####################################
#
#DATABASE COMMANDS
#
#
####################################


---------------------------------------------------------------------
DBLOGIN

Use DBLOGIN to establish a database connection through GGSCI in
preparation to issue other GoldenGate commands that affect the
database.

Syntax:
DBLOGIN  
{SOURCEDB <dsn> |
USERID <user>[, PASSWORD <password>] [SYSDBA]
SOURCEDB <dsn>, USERID <user>[, PASSWORD <password>]}

SOURCEDB <dsn>
Supplies a datasource name. Required for all databases that use ODBC.

USERID <user>[, PASSWORD <password>] [SYSDBA]
Supplies a database user and that users password. Use if database
credentials are required. PASSWORD is optional. When used, the password
is echoed. If PASSWORD is omitted, GoldenGate prompts for a password,
and the password is not echoed. The SYSDBA option logs the user in as
a SYSDBA user.

Example 1: 
DBLOGIN USERID ggs, PASSWORD ggs123 SYSDBA

Example 2: 
DBLOGIN SOURCEDB CTDB@qahost, USERID ADMIN, PASSWORD ADMIN

---------------------------------------------------------------------
ENCRYPT PASSWORD

Use ENCRYPT PASSWORD to encrypt the GoldenGate database login password
that is specified with the USERID or ASMUSERID parameter. By default,
GoldenGate generates a random encryption key, but you can specify a key
from a lookup file by using the ENCRYPTKEY option.

ENCRYPT PASSWORD prints the encrypted password to the screen. Copy and
paste it into the PASSWORD argument of USERID.

Password encryption is not supported for SQL/MX databases.

Syntax:
ENCRYPT PASSWORD <password> [ENCRYPTKEY <keyname>]

<password>
The login password. Do not enclose the password within quotes.

ENCRYPTKEY <keyname>
Optional, specifies the logical name of an encryption key contained in
the ENCKEYS lookup file. GoldenGate uses the key name to look up the
actual key in the file.
To use the <keyname> option, generate the key and store it in an
ENCKEYS lookup file on the local system. For more information, see the
GoldenGate for Windows and UNIX Administrator Guide.

Example: ENCRYPT PASSWORD ny14072 ENCRYPTKEY superkey2

---------------------------------------------------------------------
LIST TABLES

Use LIST TABLES to list all tables in the database that match the
specification provided with the command argument. Use the DBLOGIN
command to establish a database connection before using this command.

Syntax:
LIST TABLES <table>

<table>
The name of a table or a group of tables specified with a wildcard
(*).

Example: The following shows a LIST TABLES command and sample output.

GGSCI (sysa) 3> list tables tcust*
TCUSTMER
TCUSTORD

---------------------------------------------------------------------
####################################
#
#MISCELLANEOUS COMMANDS
#
#
####################################


---------------------------------------------------------------------
HELP

Use HELP to obtain information about a GoldenGate command. The basic
command returns a list of categories and their commands. The
<command> option restricts the output to that of a specific command.

Syntax:
HELP [<command>]

<command>
The command for which you want help.

Example: HELP add replicat

---------------------------------------------------------------------
HISTORY

Use HISTORY to view a list of the most recently issued commands in
GGSCI since the session started.

Syntax:
HISTORY [<n>]

<n>
Returns the last <n> commands, where <n> is any positive number.

Example:
HISTORY 3

---------------------------------------------------------------------
INFO ALL

Use INFO ALL to display the status and lag (where relevant) for all
Manager, Extract, and Replicat processes on a system. The basic
command, without options, displays only online (continuous) processes.
To display tasks, use either INFO ALL TASKS or INFO ALL ALLPROCESSES.
 
Syntax:
INFO ALL [TASKS | ALLPROCESSES]

TASKS	
Displays information only for tasks.

ALLPROCESSES
Displays information for online processes and tasks.

Example 1: INFO ALL TASKS
Example 2: INFO ALL ALLPROCESSES

-------------------------------------------------------------------
INFO MARKER

Use INFO MARKER to review recently processed markers from a NonStop
system. A record is displayed for each occasion on which GGSCI, Logger,
Extract, or Replicat processed the marker.

Markers can only be added on a NonStop system, using GoldenGate for
HO NonStop software.

Syntax:
INFO MARKER [COUNT <num items>]

COUNT <num items>
Restricts the list to a specified number of the most recent markers.

Example: INFO MARKER COUNT 2

--------------------------------------------------------------------
OBEY

Use OBEY to process a file that contains a list of GoldenGate 
commands. This command is useful for executing commands that are
frequently used in sequence.

Syntax:
OBEY <file name>

<file name>	
The fully qualified name of the file containing the list of commands.

Example: OBEY ./mycommands.txt

---------------------------------------------------------------------
SHELL

Use SHELL to execute shell commands from within the GGSCI interface.

Syntax:
SHELL <command>

<command>
The system command to execute.

Example 1: SHELL dir dirprm\*
Example 2: SHELL rm ./dat*

---------------------------------------------------------------------
SHOW

Use SHOW to display the current GoldenGate environment.

Syntax:
SHOW

---------------------------------------------------------------------
VERSIONS

Use VERSIONS to display operating system and database version
information.  For ODBC connections, driver version information is
displayed as well. To display database information, issue a DBLOGIN
command first to establish a database connection.

Syntax:
VERSIONS

---------------------------------------------------------------------
VIEW GGSEVT

Use VIEW GGSEVT to view the GoldenGate error log (ggserr.log file).
This file contains information about GoldenGate events, such as
process startup, shutdown and exception conditions. This information
also is recorded in the system error log, but viewing the GoldenGate 
error log is sometimes more convenient and may retain events further
back in time.

The display can be lengthy. To exit the display before reaching
the end, use the operating system's standard methods for terminating
screen output.

Syntax:
VIEW GGSEVT

---------------------------------------------------------------------
VIEW REPORT

Use VIEW REPORT to view the process report that is generated by 
Extract or Replicat. The report lists process parameters, run
statistics, error messages, and other diagnostic information.

The command displays only the current report. Reports are aged
whenever a process starts. Old reports are appended with a sequence
number, for example finance0.rpt, finance1.rpt, and so forth. To view
old reports, use the [<n>] option.

Syntax:
VIEW REPORT {<group name>[<n>] | <file name>}

<group name>
The name of the group. The command assumes the report file named
<group>.rpt in the GoldenGate dirrpt sub-directory.

<n>
The number of an old report. Report files are numbered from 0 
(the most recent) to 9 (the oldest).

<file name>
A fully qualified file name, such as c:\ggs\dirrpt\orders.rpt.

Example 1: VIEW REPORT orders
Example 2: VIEW REPORT orders3
Example 3: VIEW REPORT c:\ggs\dirrpt\orders.rpt

---------------------------------------------------------------------
!

Use ! to execute the previous GGSCI command without modifications. To
modify a command before re-executing it, use the FC command.

Issuing ! without arguments re-executes the most recently used command.
By using options, you can re-execute a specific command by specifying
its line number or a text substring. Previous commands can only be 
re-executed for the current session of GGSCI, because command history
is not maintained from session to session.

Syntax:
! [<n> | -<n> | <string>]

<n>
Re-executes the command from the specified GGSCI line. Each GGSCI 
command
line is sequenced, beginning with 1 at the start of the session.

-<n>
Re-executes the command issued <n> lines before the current line.

<string>
Re-executes the last command that starts with the specified text
string.

Examples:
! 9
! -3
! sta


----------------------------------------------------------------------
FC

Use FC to edit a previously issued GGSCI command and then re-execute
it. Previous commands are stored in the memory buffer. Issuing FC
without arguments re-executes the most recently used command. By using
options, you can retrieve a specific command by specifying its line
number or a text substring. Previous commands can only be edited for
the current session of GGSCI, because command history is not maintained
from session to session.

The FC command displays the command and then opens an editor with a
prompt containing a blank line starting with two dots. To edit a
command, use the space bar to position the cursor beneath the character
in the displayed command where you want to begin editing, and then
enter one of the following arguments. Arguments are not case-sensitive
and can be combined.

     i <text>
     Inserts text. 

     r <text>
     Replaces text.

     d
     Deletes a character. To delete multiple characters, enter a d for
     each one.

To execute the command, press Enter twice, once to exit the editor and
once to issue the command. To cancel an edit, type a forward slash (/)
twice.

Syntax:
FC [<n> | -<n> | <string>]

<n>
Returns the command from the specified GGSCI line. Each GGSCI command
line is sequenced, beginning with 1 at the start of the session. 

-<n>
Returns the command that was issued <n> lines before the current line.

<string>
Returns the last command that starts with the specified text string. 

Examples:
FC 9
FC -3
FC sta

---------------------------------------------------------------------
####################################
#
#CHECKPOINTTABLE COMMANDS
#
#
####################################


----------------------------------------------------------------------
ADD CHECKPOINTTABLE

Use ADD CHECKPOINTTABLE to create a checkpoint table in the target
database. Replicat uses the table to maintain a record of its read
position in the trail for recovery purposes.

A checkpoint table is optional; checkpoints are also maintained in a
file on disk. The use of a checkpoint table causes checkpointing to be
part of the Replicat transaction. This system allows Replicat to
recover in certain circumstances more seamlessly than if just file
checkpoints alone are used.

One table can serve as the default checkpoint table for all Replicat
groups in a GoldenGate instance if you specify it with the
CHECKPOINTTABLE parameter in a GLOBALS file. More than one instance of
GoldenGate (multiple installations) can use the same checkpoint table.
GoldenGate keeps track of the checkpoints, even if Replicat group names
are the same in different instances. For more information, see the
GoldenGate for Windows and UNIX Operations Guide.

Use the DBLOGIN command to establish a database connection before using
this command. Do not change the names or attributes of the columns in
this table. You may, however, change table storage attributes.

Syntax:
ADD CHECKPOINTTABLE [<owner.table>]

[<owner.table>]
The owner and name of the checkpoint table to be created. Use this 
argument to create a non-default table that will be specified with the 
CHECKPOINTTABLE argument of ADD REPLICAT. 

It is recommended, but not required, that the table be created in a 
schema dedicated to GoldenGate. If an owner and name are not specified, 
a default table is created based on the CHECKPOINTTABLE parameter in 
the GLOBALS parameter file. 

Record the name of the table, because you will need it to view 
statistics or delete the table if needed.

Example 1: ADD CHECKPOINTTABLE
Example 2: ADD CHECKPOINTTABLE ggs.fin_check


----------------------------------------------------------------------
CLEANUP CHECKPOINTTABLE

Use CLEANUP CHECKPOINTTABLE to remove checkpoint records from the
checkpoint table when there is no checkpoint file associated with it in
the working GoldenGate directory (from which GGSCI was started). The
purpose of this command is to remove checkpoint records that are not
needed any more, either because groups were changed or files were
moved.

Use the DBLOGIN command to establish a database connection before using
this command.

Syntax:
CLEANUP CHECKPOINTTABLE [<owner.table>]

[<owner.table>]
The owner and name of the checkpoint table to be cleaned up. To 
preserve case sensitivity, enclose the names within quotes. If an owner
and name are not specified, the table that is affected is the one
specified with the CHECKPOINTTABLE parameter in the GLOBALS parameter
file.

Example 1: CLEANUP CHECKPOINTTABLE
Example 2: CLEANUP CHECKPOINTTABLE ggs.fin_check


----------------------------------------------------------------------
DELETE CHECKPOINTTABLE

Use DELETE CHECKPOINTTABLE to drop a checkpoint table from the
database. Use the DBLOGIN command to establish a database connection
before using this command.

To stop using a checkpoint table while the associated Replicat group
remains active, follow these steps:

1. Run GGSCI.

2. Stop Replicat.
     STOP REPLICAT <group>

3. Delete the Replicat group and then add it back with the option
   shown.
     DELETE REPLICAT <group>
     ADD REPLICAT <group>, EXTTRAIL <trail>, NODBCHECKPOINT

4. Exit GGSCI, then start it again.

5. Start Replicat again.
     START REPLICAT <group>

6. Log into the database with DBLOGIN, using any of the options shown,
   depending on the authentication credentials required.
   
   DBLOGIN  
   {SOURCEDB <dsn> |
   USERID <user>[, PASSWORD <password>] |
   SOURCEDB <dsn>, USERID <user>[, PASSWORD <password>]}

7. Delete the checkpoint table with DELETE CHECKPOINTTABLE.

If the checkpoint table is deleted while Replicat is still running and
transactions are occurring, Replicat will abend with an error stating 
the checkpoint table could not be found. However, the checkpoints are 
still maintained on disk in the checkpoint file. To resume processing, 
add the checkpoint table back under the same name. Data in the trail 
resumes replicating. Then, if you still want to delete the checkpoint 
table, follow the preceding steps for deleting the table.

Syntax:
DELETE CHECKPOINTTABLE [<owner.table>] [!]

<owner.table>
The owner and name of the checkpoint table to be deleted. An owner and
name are not required if they are the same as those specified with the
CHECKPOINTTABLE parameter in the GLOBALS file.

!
Bypasses the prompt that confirms intent to delete the table.

Example: DELETE CHECKPOINTTABLE ggs.fin_check


---------------------------------------------------------------------
INFO CHECKPOINTTABLE

Use INFO CHECKPOINTTABLE to confirm the existence of a checkpoint table
and view the date and time that it was created. It returns a message
similar to the following:

Checkpoint table HR.CHKPT_TBLE created 2005-10-06 11:51:53.

Use the DBLOGIN command to establish a database connection before using
this command.

Syntax:
INFO CHECKPOINTTABLE [<owner.table>]

<owner.table>
The owner and name of the checkpoint table. An owner and name are not
required if they are the same as those specified with the
CHECKPOINTTABLE parameter in the GLOBALS file.

Example:
INFO CHECKPOINTTABLE ggs.fin_check

---------------------------------------------------------------------
####################################
#
#DDL COMMANDS
#
#
####################################


---------------------------------------------------------------------
DUMPDDL

Use the DUMPDDL command to view the data in the GoldenGate DDL history
table. This information is stored in proprietary format but can be
exported in human-readable form to the screen or to a series of SQL
tables that can be queried by using regular SQL.

The information in the history table is the same information that is
used by the Extract process. Because this information is historical
data that is provided by the DDL before trigger (default name of
GGS_DDL_TRIGGER_BEFORE), it reflects the state of an object before a
DDL change. Consequently, there will not be any data for CREATE
operations.

To use DUMPDDL, you must first log into the database as the owner of
the history table by using the DBLOGIN command.

The basic DUMPDDL command outputs metadata to a set of tables whose
names begin with "GGS_DDL_". The SEQNO column is the primary key of the
main table, the GGS_DDL_OBJECTS table, and the other tables contain the
same column as a foreign key. SEQNO is the DDL sequence number that is
listed in the Extract and Replicat report files. It also can be
obtained by querying the DDL history table (default name is 
GGS_DDL_HIST).

All of these tables are owned by the schema that was designated as the 
GoldenGate DDL schema during the installation of the DDL objects. To 
view the structure of these tables, use the DESC command in SQL*Plus.

Additional DUMPDDL guidelines:

DUMPDDL always dumps all of the records in the DDL history table. Use
SQL queries or search redirected standard output to view information
about particular objects and the operations you are interested in.
Because the history table contains large amounts of data, only the
first 4000 bytes (approximately) of a DDL statement are displayed in
order to maintain efficient performance.

The format of the metadata is string based. It is fully escaped and
supports non-standard characters (such as =, ?, *) in table or column
names.

Syntax: DUMPDDL [SHOW]

SHOW
DUMPDDL with the SHOW option dumps the information contained in the
history table to the screen in standard output format. No output tables
are produced. All records in the DDL history table are shown.
