* copyright (C) 1984-2019 merrill consultants dallas texas usa

variables in the CICSACCT By list, below.
The following list shows the original sort order with

a + to indicate where new sort variables were added.

Dataset BY Variables

CICSACCT APPLID+ TERMINAL OPERATOR USER TRANTYPE TRANNAME ENDTIME

CICSEXCE APPLID+ TERMINAL OPERATOR USER TRANTYPE TRANNAME ENDTIME

CICSYSTM APPLID STRTTIME+ ENDTIME

TYPE0 SYSTEM+ IPLTIME

TYPE21 SYSTEM+ SMFTIME DEVNR

TYPE70 SYSTEM STARTIME+ SMFTIME

TYPE71 SYSTEM STARTIME+ SMFTIME

TYPE72 SYSTEM STARTIME+ PERFGRP PERIOD SMFTIME

TYPE73 SYSTEM STARTIME+ CHPID SMFTIME

TYPE73P SYSTEM STARTIME+ CHAN SMFTIME

TYPE73L SYSTEM STARTIME+ LCHAN SMFTIME

TYPE74 SYSTEM STARTIME+ DEVNR SMFTIME

TYPE75 SYSTEM STARTIME+ DEVNR PAGEDSN SMFTIME

TYPE78 SYSTEM STARTIME+ LCUID SMFTIME

TYPE78CF SYSTEM STARTIME+ LCUID CHPID SMFTIME

TYPE78CU SYSTEM STARTIME+ LCUID SMFTIME

TYPE78IO SYSTEM STARTIME+ IOPIQID SMFTIME

TYPE78PA SYSTEM STARTIME+ READTIME JOB SMFTIME

TYPE78SP SYSTEM STARTIME+ READTIME JOB SMFTIME

TYPE78VS SYSTEM STARTIME+ SMFTIME

TYPE30_D READTIME JOB JESNR+ INITTIME INTBTIME SMFTIME

TYPE30_V READTIME JOB JESNR+ INTBTIME INTETIME

TYPE30_1 READTIME JOB JESNR JINTTIME+

TYPE30_4 READTIME JOB JESNR TERMTIME+

TYPE30_5 READTIME JOB JESNR DESCENDING JTRMTIME+

TYPE6 READTIME JOB JESNR PRINTIME+ PRENTIME OUTDEVCE SYSTEM

TYPE26 READTIME JOB JESNR+ JPURTIME SYSEXEC

Thanks to ???, NRMA, AUSTRALIA.

who suggested this in 1984.

Change 04.37 Variable EXCPRMF in type 30 data sets is calculated

Mar 26, 1986 from data in the 30 (IOUNITS) and from the IOCCOEFF

VMAC30 coefficient in the type 72. If a 30 was encountered

VMAC7072 before a 72, EXCPRMF would be missing. However, if

duplicate 30 records were separated by a 72, the

first 30 had missing EXCPRMF, while the duplicate had

a value, which prevented NODUP option in SORT from

recognizing the true duplicate. Now, IOCSYSTM is

retained from 72 and compared with SYSTEM in 30 to

only create EXCPRMF when preceding 72 is from same

system, and thus allow NODUP option to delete dupes.
Change 04.36 When the EJST,SRBT suffix variables from QWSA were

Apr 07, 1986 split into QWS1 and QWS2, and when ISEQ,SBNA,SCF,

ANALDB2 SRND,SRNW,SRSW,OTH1,OTH2 suffix variables from QWSC

were split into QWS1, QWS2 and QWS3 prefixes, the

DIF() operations necessary to make the DB2 data

meaningful were overlooked in ANALDB2, which affects

those variables (and CPUTM) in DB2STAT0 data set.

New variables DB2TCBTM,DB2SRBTM,ELAPSTM,JOB are now

created in DB2ACCT.

Remember, ANALDB2 is not optional; it is automatically

invoked by TYPEDB2, and must be included in EXPDBCDE

if you create DB2 data sets with BUILDPDB.

Thanks to Mike Schilling, NORWEST Information Services, USA.

Change 04.35 TERMADDR and MINIADDR in several VM data sets built

Mar 24, 1986 from VM Accounting Card data should not be $HEX6. as

TYPEVM they were characters all along. Remove from FORMAT.

Thanks to ???, ???.
Change 04.34 Under XA only, the PAG values in SMF are the sum of

Mar 24, 1986 PAG + FIX, but MXG did not recognize that fact. With

VMAC71 this change, the MXG value of PAG variables will be

correct (and are less than the MXG V3 values).

Thanks to Les Czegel, CIL, CANADA.
Change 04.33 Interval begin time INTBTIME added to the TYPE30_D DD

Mar 24, 1986 level data set, so that DD observations from different

VMAC30 intervals of the same step (i.e., same INITTIME) can

be recognized.

Thanks to Chuck Hopf, Computer Language Research (FASTAX), USA.
Change 04.32 1. The Vector Processor accounting times (VPU & VPI) for

Mar 24, 1986 USE and AFFinity were added to PDB.JOBS and PDB.STEPS.

BUILDPDB 2. INTBTIME (interval begin time) was added to the BY

BUILDPD3 list (READTIME JOB JESNR INTBTIME INTETIME) for

TYPE30_V (now that RMF 3.3 provides INTBTIME).

4. SMFTIME was appended to the BY list (READTIME JOB

JESNR INITTIME SMFTIME) for TYPE30_D (DD statistics)

so DD records from different intervals (which will

have same INITTIME) can be identified and ordered.
Change 04.31 New data in TYPE21 record now supported. New counts

Apr 4, 1986 of errors for 3480, tape unit serial number, DCBOFLG,

VMAC21 and separate counts of bytes written and read added.

See TNL GN28-1095 for PTF OZ89909, Jan 16, 1986.

Change 04.30 Data sets TYPE78PA, TYPE78SP, and TYPE78VS were not

Mar 20, 1986 created by BUILDPDB (but were for BUILDPD3). This

BUILDPDB change adds those three (small) data sets to the PDB.

Thanks to Wing Louie, Metropolitan Life, USA.

Change 04.29 ZDATE ("Zee date zee data was added to zee PDB") was

Mar 20, 1986 added to _JOURNAL macro for CICS journal format data.

RMFINTRV ZDATE was also added to the KEEP list for RMFINTRV.

VMACSMF

Thanks to Cathy Hellums, Aristar, USA.
Change 04.28 TYPE72 and TYPE74 data sets do not always contain all

Mar 20, 1986 possible observations. Only if SERVICE or SIOCOUNT was

VMAC7072 non-zero would MXG OUTPUT, and the control logic was

EXTY72 in the VMAC member. Now, the control logic for their

VMAC74 OUTPUT statement has been moved to their EXTYnn member

EXTY74 so that sites which want all observations can make

the change in their exit code. Sites which use TYPE74

for availability want all observations, and sites that

use report perf. groups but capture only transactions

may want all TYPE72 observations. The default was not

changed, only the location of the default logic was.

Thanks to Barbara Watters, EXXON Houston, USA

Change 04.27 PRQUETM is now correct for JES3. For JES2, the start

Mar 20, 1986 of print queue time is JFINTIME, but JES3 does not

BUILDPD3 terminate output processing (JFINTIME) until purge.

Thus JPRNTIME (start of output processing) must be

uses for JES3 start of print queue time.
Change 04.26 4381-III CPU factor (pre-RMF 3.3) for single CPU is

Mar 20, 1986 now correct. 4381 de-rates UP by 85% when MP'ed, but

VMAC7072 all other MVS machines de-rate 95%, 92% and 89% when

2, 3, or 4 processors are online.

Thanks to David Belsham, Lloyds Corporation.
Change 04.25 Inconsistent RMF data caused NOTSORTED error in one

Mar 20, 1986 site because BY statement was missing. Additional

RMFINTRV check added to advise if RMF data was inconsistent.

Thanks to Pat Goetzinger, Northwestern Bell (of Nebraska!), USA.

Change 04.24 Production version of RMF 3.3 relocated LCUID field

Mar 4, 1986 in TYPE78 data set (only for 3090 CPUs).

VMAC78

Thanks to Greg Scriba, First National of Chicago, USA.

-----Changes thru 4.23 completed first PRE-RELEASE of Version 4.0-----
Change 04.23 Partitioning 3084s into 3081s, etc., caused MXG to

Jan 31, 1986 incorrectly calculate PCTCPUBY after splitting the

VMAC7072 CPUs. Beautiful fix (print of source, highlighted

with his changes, and with two pages of descriptive

documentation) was supplied by, and therefore

Thanks to Stephen Secher-Jensen, ANZ Bank, Melbourne, AUSTRALIA.

Change 04.22 The current version of MODEL 204 now contains the

Jan 27, 1986 M24ACCT and M24USER by default, so the exit is now

EXM24ACT set to default ON (was OFF). See member.

Thanks to David Daner, Sun Company, USA.

Change 04.21 IMS Log processing code. IMSTRAN and IMSMISS data sets

Jan 22, 1986 are built from sequences of IMS log records. This code

TYPEIMS was originally written and tested in Europe. It seems

IMACIMS to handle IMS 1.3 quite well, and is self-documenting.

EXIMSTRN Wherever possible, variable names of IMS Log data are

EXIMSMIS the same as those used for the IMF/Control IMS data

set CIMSTRAN, page 413.

Thanks to Allan Russell, SAS Institute Europe, GERMANY.

Change 04.20 JES3 BUILDPDB ABEND and CONDCODE "fixes" were not

Jan 22, 1986 completed with change 4.xx. Variables ABEND and

BUILDPD3 CONDCODE were added following KEEP= _PDB30_5 in

PROC SORT DATA=TYPE30_5 OUT=TYPE30_5 (KEEP= ....

Thanks to Bill Cohen, Drexel, Burnham, Lambert, USA.
Change 04.19 The catalog record at the end of the 61,65 and 66

Jan 15, 1986 records was not decoded. The program will now (if

VMAC6156 you choose) decode and print (but not add to the

TYPE6156 data set) these catalog segments. See the

'how to' comments in the member itself.

Thanks to Gary Bortner, Lucky Stores, USA.

Thanks to John Lemkelde, Pennsylvania Blue Shield, USA.
Change 04.18 Added five more work groups OTH1-OTH5 to RMFINTRV

Jan 8, 1986 data set. Added Exit EXRMFINT so you can label the

RMFINTRV variables to describe that OTH1 is IDMS, etc.

EXRMFINT Comments added to IMACWORK to show how to use.

IMACWORK March 20, calculation code finally added!

Thanks to George Scott, Rockwell International Corporation, USA.

Thanks to Greg Scriba, First National of Chicago, USA.
Change 04.17 Support for CICS 1.7. Major changes in internal data

Jan 8, 1986 format, with some new variables, but minimal impact

VMAC110 on your programs. Will be discussed in future issue

FORMATS of MXG Newsletter. Do not gen CICS 1.7 with EXCLUDEs.

Change 04.16 Input format for OFFDATE in DOSRJE is MMDDYY6., and

Jan 8, 1986 PRTYJOB=MOD(PRTYJOB,16) replaced PRTY=MOD(PRTY,16).

TYPEDOS

Thanks to Joseph G. Ogurchak, Westfield Companies, USA.

Change 04.15 PTF UZ90402 for RMF 3.3 alters MVS/XA RMF format for

Dec 29, 1985 Vector Processors, even if you don't have one! The

following changes were published in MXG NEWSLETTER

SEVEN prior to availability of Version 4.1. Actual

code in Version 4 is more comprehensive that the

quick fix printed in Newsletter Eight.

VMAC7072 a. VECONSAM and VECTORON variables added to TYPE70.
VMAC22 b. VECTORON added.
VMAC30 c. Step level vector utilization variables added.

VPUUSETM

VPIUSETM

VPUAFFTM

VPIAFFTM

d. Interval begin time, INTBTIME, finally put in

the subtype 2 and 3 interval records by IBM.

Thanks to George Scott, Rockwell International Corporation, USA.

Change 04.14 Constant 000000954 should be 000000953674. Causes a

Dec 29, 1985 very small rounding error. Almost a nit.

FORMATS
Thanks to Joseph J. Faska, Chemical Bank, USA.
Change 04.13 Standard deviation of response time calculation can

Dec 26, 1985 cause negative square root. Test added to avoid the

VMAC7072 illegal argument error message. SSQELAP wraps inside

RMF, creating above problem. IBM has been notified.

Change 04.12 RMF 3.3 expanded LCUID to two bytes in the subtype 1

Dec 18, 1985 record. MXG V3 missed this change, causing LCUID to

VMAC78 be zero in TYPE78CF.
Thanks to Joseph J. Faska, Chemical Bank, USA.
Change 04.11 SAS Version 5 requires quotes around all character

Dec 02, 1985 strings. LABEL and FORMAT statements had been fixed,

TYPEPRNT but execution with NOTEXT82 option suraced others.

OTHRPRNT Quotes in SPLIT='*' and quotes around hex character

FORMATS values in PROC FORMAT were added to these members:

BUILDPD3 (Also removed superfulous PROC PRINT from BUILDPD3)

ANALCICS ANALCONT ANALDALY ANALDOS ANALESV

ANALPROG ANALTSOR ANALTURN ANALVM

Thanks to David Henley, Blue Cross, USA.
Change 04.10 INTRVAVG is always missing. Change equations (two)

Dec 02, 1985 which calculate INTRVAVG to use INTRVSUM as first

VMAC76 term on right side of equation.
Thanks to Kenneth D. Jones, Maritime Telegraph and Telephone, CANADA.
Change 04.09 RESPAVG and RESPSTD and AVGMEMSZ values were carried

Dec 02, 1985 into next PERFGRP.PERIOD if TRANS=0 or CPUUNITS=0.

VMAC7072 After each of the 6 IF statements which calculate

these variables, insert 'ELSE variable=.;' to set

the value to missing.

Thanks to Sue Rosansky, Metropolitan Life, USA.

Change 04.08 Logic to set return code to 4096 or less was added

Nov 19, 1985 to parallel type 4, 30 and 34 records as well.

VMAC535

Thanks to Sam Ferrarelli, Philadelphia Electric, USA.

Change 04.07 Additional tests for zero denominator added for the

Nov 19, 1985 variable NRSAMPLE (which strangely is sometimes 0).

VMAC74

Thanks to Joseph J. Faska, Chemical Bank, USA.

Change 04.06 Change 2.22(a) was not completely applied for JES3.

Nov 19, 1985

BUILDPD3

a. Find the three lines which create TYPE30_5 from the

SET TYPE30_5 SPIN30_5; statement.

Replace those three lines with the similar eight

lines from member BUILDPDB.

b. Find MACRO _SUMSTP. Remove the variable TAPEDRVS in

that macro definition by blanking it out.

c. Find DATA ONE30_5, then find FORMAT CONDCODE HEX4.

insert two lines: ABEND=ABEND5; CONDCODE=CONDCOD5;

Thanks to George Scott, Rockwell International Corporation, USA.

Thanks to Martin Hamburg,Johns Hopkins Univ Applied Physics Lab, USA.
Change 04.05 Five new variables (RMF 3.3 only) were INPUT out of

Nov 18, 1985 order. Starting at OFFPGDS+392, variables should be

VMAC71 MIGAGEMN,MX,AV, and EXTFRMIN,ON.

Thanks to Sue Rosansky, Metropolitan Life, USA.

Change 04.04 Removed lower case from IMACINTV, as it caused SPF

Nov 6, 1985 to change case and data values became lower case!

IMACINTV

Thanks to Dennis Dwyer, CITICORP Person-to Person, USA.

Change 04.03 Minor correction to MG078CV format to add HIGH.

Nov 6, 1985

FORMATS

Thanks to Joseph J. Faska, Chemical Bank, USA.

Change 04.02 Corrected algorithm for counting tape drives used

Nov 4, 1985 and added count of tape drives potentially needed.

ANALTAPE Many lines involved. Call for information.

Thanks to Robin Templer, SAS Institute, AUSTRALIA.

Change 04.01 Backed out 2.3. Default record ID from TSO/MON and

Nov 4, 1985 MXG now agree (they're alphanumeric: CMD - SYS for

IMACTSOM 199 and 200 respectively).

Thanks to Ray Dickensheets, Yellow Freight System, USA.

LASTCHANGE: Version 4

=========================member=CHANGE03================================

/* COPYRIGHT (C) 1985 BY MERRILL CONSULTANTS DALLAS TEXAS */
MXG Software Status as of November 1, 1985
This is MXG Version 3.
This member describes the changes made to the MXG Software since

Version 2 was shipped. The documentation of new data sets and new

variables created by Version 3 is contained in member DOCVER03. The

changes previously introduced by Version 2 are in member DOCVER02.

Additional documentation is available in past and future issues of

the MXG Newsletter, sent by Merrill Consultants to their Supported

Customers. Newsletter SIX was published concurrent with Version 3.
NEXTCHANGE: Version 3
=============Changes thru 3.10 complete Version 3.2==================
Change 03.10 Format DATE9. added for DATE in RMFINTRV data set.

Oct 30, 1985

RMFINTRV

Thanks to Chuck Hopf, Computer Language Research (FASTAX), USA.

Change 03.09 Error in reading TPUTHOLD variable in TSO/MON 4.3

Oct 30, 1985

VMACTSOM

Thanks to Gayle Bartel, Pizza Hut, USA.

Change 03.08 LOGAUX... variables missing in TYPE71 with RMF 3.3

Oct 30, 1985

VMAC71

Thanks to Sue Rosansky, Metropolitan Life, USA.

=============Changes thru 3.07 were in Version 3.1===================
Change 03.07 Added format MG078CV. to map bytes to K or Meg for

Oct 18, 1985 PVTSZHI/LOW, LSQSZHI/LOW and USRSZHI/LOW. Midnight

VMAC30 logic error corrected.
Change 03.05 Error introduced in Change 2.19 caused all of the

Oct 17, 1985 TYPE70 variables .....AVG to be 100 times too big.

VMAC7072

Thanks to Chuck Hopf, Computer Language Research (FASTAX), USA.

Change 03.04 Error introduced in Change 2.24 caused loss of type

Oct 15, 1985 30 data sets observations.

VMAC30

Thanks to Roman Gudz, Leaseway, USA.

Change 03.03 IBM offset values were wrong (from start of logical

Oct 15, 1985 record) but MXG coded for it. Now IBM PSF changes

VMAC5234 for 3820 changes PROD offset, but left ID offset

still wrong. MXG now forces the OFFPROD and OFFID

values, since data location is unchanged.

Thanks to Chuck Hopf, Computer Language Research (FASTAX), USA.

Change 03.02 CICS PTF UP90227/UP90228 for APAR PP41582, an SPE

Oct 15, 1985 for CICS 1.6.1, changes CMF record version from 2

VMAC110 to 3.

Thanks to Norbert Korsche, OMV-AG, AUSTRIA.

Change 03.01 Undocumented field, QBACRIO, DB2 Synchronous Read

Oct 15, 1985 I/O requests was discovered and now in the DB2ACCT

VMACDB2 data set.

Thanks to Ron Root, Sun Company, USA.

LASTCHANGE: Version 3
NEXTCHANGE: Version 2

=============Changes thru 2.25 were in Version 3.0===================

Change 02.25 MVS 2.1.3 increased the size of MPL, WEIGHT, AOBJ,

Sep 20, 1985 DOBJ and DFWK values from one to two bytes.

TYPE90
Change 02.24 Condition code (return code) must be less than 4096.

Sep 9, 1985 Some accounting systems exploit this and use upper

TYPE30_4 bits of the two byte field. MXG decoded the entire

TYPE30_5 return code, producing invalid values. Now only the

TYPE434 rightmost bits are used for return code value.

Thanks to John Zupan, STC, USA.

Change 02.23 Variables AVG and INTRVAVG were added.

Sep 9, 1985

TYPE76

Thanks to Debby Blackey, Fidelity Savings, USA.

Change 02.22 Several enhancements made to BUILDPDB/BUILDPD3.

Sep 9, 1985 a. ABEND and CONDCODE in the JOBS data set is now the

BUILDPDB job completion status for the last job termination

BUILDPD3 (subtype 5) record. Formerly, it was the last step

termination and was misleading (i.e., ABEND would

be FLUSH for a flushed step in the JOBS dataset).

b. EXCP and IOTM for tapes 3420 and 3480 added.

c. SPINCNT (the number of times partially-completed

jobs will be held in the SPIN file awaiting their

purge record) is now set by member IMACSPIN.

d. TYPE78CF, TYPE78CU and TYPE78IO now in PDB.

e. SAS5FIX1 module created to correct the SAS error

in SAS VERSION 5 until SAS provides the first

maintenance release to VERSION 5. If you execute

BUILDPDB/BUILDPD3 under SAS VERSION 5, you must

tack SAS5FIX1 on after the PDB is built. See the

comments in SAS5FIX1.
Thanks to Malcolm Morgan, Wachovia Bank, USA.

Thanks to Georg Simon, Prudential Insurance, USA.

Change 02.21 Support for RMF Version 3.3.0, which is required for

Sep 1, 1985 3090 processors. See DOCVER03 for more details:

TYPE7072 Service unit per second constant (SU_SEC) is now in

the TYPE72 record.

TYPE71 Many new variables describing swapping between the new

extended storage and auxiliary storage, migration age

etc. Swap reasons count expanded to identify where the

various swaps are occurring.

TYPE74 Three new variable describe the components of device

pending time. Both overflow conditions (308x only)

now identified.

TYPE78CF Configuration data set now contains utilization data.

TYPE78CU New data set describes CUHDR queue length and rate.

TYPE78IO New data set describes IOP initiative queue stats.

Change 02.20 The calculation of ALOCTIME and LOADTIME in TYPE30_4

VMAC30 were corrected (as had been done in TYPE434) to add

Sep 1, 1985 a day if the job allocation/load time of day was less

than the initiate time of day.

Change 02.19 The _VAR70 ARRAY in processing type 70 records was

VMAC7072 replaced with direct assignment after a European

Sep 1, 1985 customer first noted a siginifcant increase in CPU

time of MXG TYPE70 when compared to the original

MG TYPE70, and research confirmed the ARRAY was

more CPU costly than direct assignments.

Change 02.18 New format, MG078CV. converts byte values to Kilo

FORMATS or Mega bytes, with K or M suffix. Makes the data

Sep 1, 1985 in TYPE78VS (virtual storage) easily read.

Thanks to Joseph J. Faska, Chemical Bank, USA.

Change 02.18A New data set created by RMF 3.3 for extended storage.

TYPE22_9
Change 02.17 The value of SPINCNT, which sets the number of

IMACSPIN executions that BUILDPDB will wait for a job's purge

Aug 30, 1985 record, was externalized to IMACSPIN, consistent

with moving all installation-unique values to IMAC's.
Change 02.16 The routine worked when all input was SMF, but use

ANALDSET with the PDB as input for the Step records failed

Aug 30, 1985 because of a missing RENAME.

Thanks to ???, ???.

Change 02.15 This module to process VS1 SMF records is an 'extra'

XTYPEVS1 module, undocumented and with unlabeled variables

Aug 15, 1985 but it also would not run. Removal of an unnecessary

%INCLUDE and proper calculation of NUMDD now makes it

usable for VS1 data.

Thanks to Fritz Ederer, SAS Institute, GERMANY.

and a German VS1 site.
Change 02.14 Step 3 of JCLTEST was broken into two steps, because

JCLTEST the SAS limit of 32768 SYSIN card images was exceeded.

OTHRPRNT TYPETEST/TYPEPRNT now operate on SMF records, while

OTHRTEST OTHRTEST/OTHRPRNT operate on non-SMF records. Member

PRINTING PRINTING was deleted.

TYPEPRNT

TYPETEST

Jun 30, 1985

Thanks to Joseph J. Faska, Chemical Bank, USA.

Change 02.13 Two new exits were created (for 370 or XA) to set the

IMACPUSU value of SU_SEC (CPU Factor, or the Machine Dependent

IMACPUXA Constant) when a new model CPU which is not in the

VMAC7072 table in VMAC7072 is installed. Under MVS/370, it is

Jun 30, 1985 used to separate different CPU versions.

Change 02.12 Support for Release 4.3 of the TSO/MON Product. Eight

VMACTSOM new variables: ATMAXUTM, NETDLYTM, PANELCNT, SRMTRANS,

Jun 30, 1985 TPUTHOLD, TSMSMTHK, and TSMSNMTK were added; and the

value of USRTHKTM was corrected.

Change 02.11 Added new ROSCOE monitor command AMS and protected for

VMACROSC future new command names.

Jun 30, 1985

Thanks to Chuck Hopf, Computer Language Research (FASTAX), USA.

Change 02.10 Variables RESPAVG and RESPSTD, average and standard

VMAC7072 deviation of response time per ended transaction were

Jun 30, 1985 added and SSQELAP was re-formatted and corrected. For

period 1 TSO transactions, RESPAVG is the trivial TSO

response time measurement used by IBM.

Thanks to Tom Incorvia, Macy's California, USA.

Change 02.9 LABELS were enhanced to reflect that many variables

RMFINTRV are rates per second. Labels for variables ending with

Jun 30, 1985 SWAP, TRAN, EXCP, IOTM, ACTV, RESD, and SERV now have

"per second" in their label.

Change 02.8 The test IF MCTSSDRL - COL GE 60 THEN ... should have