GLOBAL command

Usage

Use the GLOBAL command to filter input data and set default values for subcommand options. It is also used to define global processing options (such as DD statements for various data sets), the number of lines printed per page, whether the OMEGAMON for Db2® PE main internal sort occurs, and time zone adjustments for different locations.

Usage notes

There can be only one GLOBAL command in a OMEGAMON for Db2 PE execution.

The OUTCCSID (ccsid) option provides support for Db2 Unicode data conversion to MBCS EBCDIC. If specified, OUTCCSID must be the first positional parameter for the GLOBAL command.

The OMEGAMON® Db2 batch reporter employs the CUNLCNV z/OS Unicode service to perform Unicode conversion to the requested OUTCCSID(). Please see the z/OS Unicode Services User's Guide and Reference for the appropriate level of z/OS for CCSIDs supported by CUNLCNV.

Syntax

1  GLOBAL?  
2.1 %FROM/TO block
2.1  INTERVAL (
2.2.1 minutes
2.2.1! 0
2.1 )
2.1  PRESORTED (
2.2.1! NO
2.2.1 ACCEPT
2.2.1 ENFORCE
2.1 )
2.1  SPANINC (
2.2.1 NOMESSAGE
2.2.1! WARNING
2.2.1 ERROR
2.2.1 NO
2.1)
2.1  BOUNDARY (
2.2.1 minutes
2.2.1 0
2.2.1! 60
2.1 )
2.1  INPUTDD (
2.2.1! INPUTDD
2.2.1 ddname
2.1 )
2.1  PAGESIZE (
2.2.1! 70
2.2.1 integer
2.1 )
2.1  TIMEZONE (
2.2.1! 0
2.2.1   +  hh : mm  
2.2.1   -  hh : mm  
2.2.1 location
2.1 )
2.1 %EXPLAIN block
2.1 %INCLUDE/EXCLUDE block

Subcommand options

The syntax diagram shows the options that are available with this subcommand. See Subcommand options for comprehensive descriptions of these options. The following list gives additional or specific descriptions of selected options, where appropriate.

FROM/TO

The range of record timestamps processed by the primary filtering function of GLOBAL.

For details, see FROM/TO subcommand options.

INTERVAL

Defines the time interval that data is summarized for.

For more information, see INTERVAL subcommand option.

PRESORTED

Controls the internal sort. It has the following values:

NO

Does not disable the sort. This is the default.

ACCEPT

Disables the sort and accepts out-of-sequence records. By using this option, you can create complete and accurate Accounting reports (including records from several locations) from data sets that have not been sorted, for example SMF or GTF. The following limitations apply with this option:

• Duplicate input data is not checked. Therefore, do not concatenate two data sets that contain the same trace records, such as when trace data has been collected in two data sets during the same period.
• Only one location is reported per trace. To report more, specify several TRACE subcommands with INCLUDE on LOCATION.
• Trace entries might not be printed in time sequence. If a trace contains entries that are out of sequence, a message appears at the end showing the number.
• When INTERVAL(0) is in effect (the default), the interval times appearing in a report heading might not be accurate. Avoid ordering reports by interval when the interval is zero.

ENFORCE

Disables the sort but terminates processing if out-of-sequence records are present. Use this option only when the input data set has been sorted, for example when reprocessing data from the DPMOUT data set.

Note: Do not specify PRESORTED for Parallel I/O or Sysplex query parallelism, the results are unpredictable. Use the default (PRESORTED(NO) if you see message FPEA4534I.

SPANINC

Controls how underlying sort facility of OMEGAMON for Db2 PE responds to incomplete, spanned records in variable-length, unblocked, spanned (RECFM=VS) or variable-length, blocked, spanned (RECFM=VBS) records.

Sort facilities like DFSORT™ provide options to control the action to be taken if incomplete spanned records are detected. For DFSORT, the option SPANINC=value, where value is RC0, RC4, or RC16, controls the return code to be set, the message to be issued, and whether to terminate if necessary. For more details, see the z/OS® information in IBM® Documentation.

By default, OMEGAMON for Db2 PE passes option SPANINC=RC4 to the underlying sort facility. For DFSORT, this option sets a return code of 4, issues message ICE1971 once, eliminates all incomplete spanned records, and continues with valid records being recovered. Usually, if no errors are detected, this process is transparent.

The following GLOBAL SPANINC subcommand options can be used to control the underlying sort facility if the default is not sufficient. The described behavior only applies to DFSORT and functionally equivalent sort facilities.

NOMESSAGE

Corresponds to SPANINC=RC0. Sets a return code of 0, issues message ICE197I (once), and eliminates all incomplete spanned records it detects.

WARNING

Corresponds to SPANINC=RC4. Sets a return code of 4, issues message ICE197I (once), and eliminates all incomplete spanned records it detects.

ERROR

Corresponds to SPANINC=RC16. Sets a return code of 16, issues message ICE204A, and terminates if an incomplete spanned record is detected.

NO

No SPANINC option is passed to the underlying sort facility. Use this subcommand option if you run a sort facility that does not support the SPANINC option or the listed option values.

BOUNDARY

Controls the alignment of the intervals used to summarize records in the reduction process.

For more information, see BOUNDARY subcommand option.

INPUTDD

The ddname of the input data set. The default is INPUTDD.

SYSPRMDD

The ddname for the System Parameters report. The default is SYSPRMDD.

PAGESIZE

The number of lines printed per page. Specify an integer in the range 50 to 999. The default for PAGESIZE is 70.

Some reports have a fixed number of lines per page. PAGESIZE is ignored for I/O Activity summary reports, multi-page records in long Record Trace reports, and Explain reports.

TIMEZONE

The time adjustment applied to record timestamps during OMEGAMON for Db2 PE processing. By using TIMEZONE, you can process data from locations in different time zones based on the local time of a single location. All further processing is based on the adjusted time. The timestamp used in FROM and TO, printed on reports and traces, and recorded in the file and save data sets is the adjusted value. The DPMOUT data set contains both the original and the adjusted value.

Note:

When combining newly reduced data with restored data, make sure that the TIMEZONE specifications for the new data match those for the restored data. Misleading results can occur if the time adjustments are different.

If you change the reference location for data, the time adjustments in old save or file data might be incompatible with newly processed data.

This does not apply to the DPMOUT data, as time adjustments are recalculated when the data is read from INPUTDD.

You can specify time adjustments as follows:

location

The location of the reference time zone. The time is adjusted relative to the specified location. The adjustment is the difference between the time zone of the reference location and the CPU clock of the reported location. The LOCDATA member of the DPMPARMS data set must contain an entry for this location.

This is the recommended method.

+hh:mm

-hh:mm

The time difference between the reference time zone and Greenwich Mean Time (GMT). The time difference is specified as ±hh:mm, where hh is hours in the range 00 to 23, and mm is minutes in the range 00 to 59. You can specify any value in the range -12:00 to +12:00. Use + for local times west of Greenwich, and - for local times east of Greenwich. The time is adjusted relative to the specified time zone difference.

The data for calculating the required adjustments is stored in the LOCDATA member of the DPMPARMS data set. It is stored and edited by using the IRF.

Note:

1. If TIMEZONE is not specified, no timestamps are adjusted. If the location supplied in the TIMEZONE option does not have a corresponding entry in the LOCDATA member, a message is generated and execution is terminated.
2. During OMEGAMON for Db2 PE processing, if data is encountered for a location whose time zone adjustment values have not been supplied in the LOCDATA member, the time adjustment default is applied to all records for that location. The default adjustment is stored in LOCDATA under a location name of *. If LOCDATA does not contain a default entry, no adjustment is applied.
3. If several systems that should have the same CPU clock time have synchronization errors, you can use TIMEZONE to correct the times for OMEGAMON for Db2 PE processing. Change the CPU clock time for the nonsynchronized locations in LOCDATA to reflect the errors. For example, if SAN_JOSE_LAB is one minute late in synchronizing with SANTA_TERESA_LAB, edit LOCDATA and add a minute to the CPU clock value for SAN_JOSE_LAB.
4. Synchronization can only be adjusted to the nearest minute. Any adjustments made to LOCDATA might become invalid if the CPU clock time at any of the locations is reset.

For more information about TIMEZONE, see Monitoring Performance from ISPF..

EXPLAIN

Defaults for EXPLAIN options can be specified with the GLOBAL command. These defaults are overridden by specifying different values in the EXPLAIN command, see GLOBAL command with EXPLAIN option.

INCLUDE/EXCLUDE

Includes or excludes data associated with specific OMEGAMON for Db2 PE identifiers.

Table 1 shows which OMEGAMON for Db2 PE identifiers can be used with the different INCLUDE and EXCLUDE subcommand options for GLOBAL.