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

One instance caused A NUMBER HAS BECOME TOO LARGE error.
==Changes thru 13.046 were contained in MXG 13.01 dated Apr 27, 1995==
Change 13.046 Report enhancements to the CICS Shutdown reports, titles

ANALCISH for detail reports, and other cosmetics were added by

Apr 26, 1995 this extensive user contribution.

Thanks to Willi Weinberger, IDG, Germany.

Change 13.045 LENGTH statement was moved to before FORMAT statement so

IMACICSA character variables' length is correct. See SAS Technical

Apr 25, 1995 Note (above, or in MXG Newsletter TWENTY-EIGHT).

Thanks to Stefaan Lemaire, COMPAREX Information Systems, BELGIUM.

Change 13.044 Missing value for timestamp was not protected, causing

ANALCNCR invalid DO loop control variable. ASUMTALO now throws

ASUMTALO away invalid records, and ANALCNCR is now protected for

Apr 25, 1995 missing value in DO loop control variables.

Change 13.043 These members were revised to work with XMXGSUM. They had

ASUMVMON exploited a loophole in VMXGSUM (the right side of the

TRNDVMON NORM= list was an expression instead of a variable name)

Apr 25, 1995 that does not exist in XMXGSUM.

Change 13.042 -The mapping of DB2 DBID/OBID hex values to the table name

ANALDB2R uses the IFCID 105 and 107 records to dynamically create

Apr 25, 1995 a look up table, the $DB2DBID and $DB2OBID formats. The

logic assumed a one-to-one mapping, but it turns out the

values get reused (DELETE and then CREATE a tablespace)

so the logic for creating these formats was redesigned to

append the timestamp of the 105/107 to create the range

of datetimes when that DBID/OBID pair was in effect.

-ANALDB2R now will execute under OS2 - the logic to detect

if the PDB pointed to an MVS tape failed on a non-MVS

platform. Also, OPTIONS LINESIZE=132 is now specified so

proper print width is used for these reports.

Thanks to ???, ???, USA.
Change 13.041 TYPE72MN variables WSETFIX/WSETASM were incorrect.

VMAC7072 They are now created using:

Apr 25, 1995 WSETFIX=FRAMEFIX/AVGUSER;

WSETASM=FRAMEASM/AVGUSER;

Thanks to Jim Gill, St. Johns River Power Park, USA.
Change 13.040 Dataset DB2ACCT variable QBACSWS and QBnCSWS were always

VMACDB2 zero; statement (QBACPID QBACGET QBACRSVD QBACSWS ...

Apr 25, 1995 should have been (QBACPID QBACGET QBACSWS QBACRSVD ...

Thanks to Jason Lau, CSC, AUSTRALIA.

Change 13.039 -The SAS/CPE product expects a "system" variable in each

DIFFTMDB dataset built by MXG, and their cross-check uncovered

TYPEMON8 these datasets which did not have a "system" variable:

TYPETMON MONIDSA - SYSTEM (retained from "TI" record)

VMACCIMS TYPECTLD - SYSTEM (was missing from KEEP= list)

VMACCTLD TMDBDE - SYSTEM (retained from "TA" record)

VMACDCOL -Dataset labels were missing in CIMSxxxx (IMF) datasets.

VMACOMSM -Variable DCDRECFM has been created in dataset DCOLDSET

VMAC7072 -SAS/CPE also expects a "duration" variable in interval

VMAC89 datasets; most MXG interval datasets contain STARTIME,

Apr 25, 1995 ENDTIME, and DURATM, but there were some omissions that

VMAC28 have now been corrected for consistency; you should now

May 2, 1995 expect those three variables in all interval datasets.

TMDBDA - No duration in record, required new member

DIFFTMDB to be invoked by TYPETMDB to SORT

and then create DURATM=DIF(ENDTIME);

OMSMSJOB - DURATM=OMFS2ETM-OMFS2STM;

TYPE7204 - DURATM should have been in KEEP= list.

TYPE7204 - DURATM should have been in KEEP= list.

TYPE89 - DURATM=SMF89IET-SMF89IST;

NPMxxxxx - DURATM=ENDTIME-STARTIME added to datasets

NPMNCINT NPMNGINT NPMNLINT NPMNLSUM NPMNRINT

NPMVSVAD NPMVSVAP NPMVSVBF NPMVSVDV NPMVSVVR

NPMNVINT NPMNVSUM NPMINTRI NPMINNEO NPMIN25L

NPMIN25P NPMINNVC NPMINFRP NPMLANIN

Thanks to Carl Sommer, SAS Institute Cary, USA.

Change 13.038 Support for subtypes 4 and 5 of HMF SMF record was added

EXTYHMFR to decode both subtypes; the subtype 4 creates a new data

IMACHMF set, TYPEHMFR, with path/route detail, while TYPEHMF4 has

VMACHMF the node statistics data. Subtype 4 has been data-tested

Apr 25, 1995 but there were no subtype 5s in the test data file.

Thanks to Shaheen Pervaiz, Acxiom, USA.

Change 13.037 Reports showing union of all systems were missing from

ANALTAPE the revision introduced in MXG 12.12.

Apr 21, 1995

Thanks to ???, ????, USA.

Change 13.036 Continued validation of ANALCNCR closed these exposures:

ANALCNCR -GOPTIONS HTITLE=1 HTEXT=.75 added to reduce the size of

Apr 21, 1995 text and titles to readable proportions.

-Plot of summary had MAX axis value slightly lower than

the max value plotted, and DEVICE= specification was

ignored.

-Plots failed with 0 observations in datasets.

-Invalid DO loop range when no SUMBY= was specified and

intervals were being synchronized.

Thanks to Tom Parker, Hogan Systems, USA.

Change 13.035 Support for APAR OW04653 which adds several variables to

VMAC74 TYPE74ST (Coupling Facility Structures) which appear to

Apr 20, 1995 be quite important in tracking capacity of the Coupling

Facility:

R744SCUF R744SDEC R744SDEL R744SMAE R744SMAS R744SMIS

R744SNLH R744SSIZ

These changes and the MXG corrections in Change 13.004

are important enough for users of MVS/ESA 5.1 in Goal

Mode (Workload Manager) that MXG 13.01 is required.

Users with MVS/ESA 5.1 still in compatibility mode still

require only MXG 12.12, but installing 13.01 is still

recommended.

Thanks to Scott Coryell, First Bank System, USA.
Change 13.034 Protection for RACFTYPE=55 to prevent "SKIPPED SEGMENT"

VMAC80A message, and variable RACFEVNT= was added to the SKIPPED

Apr 19, 1995 SEGMENT message to identify which event contained that

segment, in case it is necessary to decode and add any

new variables.

Thanks to Henrik Tonnisen, Jyske Bank Data A/S, DENMARK.

Change 13.033 A new global macro variable, MXGDEBUG, is now defined in

VMXGINIT VMXGINIT, to facilitate internal debugging %MACRO code.

Apr 18, 1995 Globaling this macro name allows debugging to be enabled

without source code changes; however, I expect that this

option will typically be used only by MXG Tech Support in

problem resolution, or during developmental testing.

It is currently implemented only in member XMXGSUM, with

these initial arguments defined:

1= Display normally non-printed messages about which

variables are being dropped because they were not

found in the input datasets - VMXGSUM specific.

2= Turn on MPRINT

3= Turn on MPRINT SYMBOLGEN

4= Turn on MPRINT SYMBOLGEN MLOGIC

You would use %LET MXGDEBUG=x; to enable this debug.
Change 13.032 More validation of this soon-to-be-replacement for the

XMXGSUM VMXGSUM member. If a variable appeared twice in one of

Apr 21, 1995 the lists or is not found in the input dataset, the

wrong number of 'Xn' variables to skip over in the

PROC MEANS was calculated, causing output variables to

contain incorrect values. The logic to calculate how

many variables are to be skipped was corrected.

-Ranges of variables BKT1-BKT8 were supported, but parsing

BKT1-BKT18 missed variables 9-18; that is now corrected.

-If all variables are missing for a NORM parameter, a note

is printed and the parameter is ignored.

Thanks to Diane Eppestine, Southwestern Bell, USA.

Thanks to Tom Elbert, John Alden, USA.
Change 13.031 Variables QTPUBDD and QTXAIRLM were incorrectly spelled

TRNDDB2S as QTPUBD and QTXAIRL respectively in MXG 12.12/12.12A.

Apr 18, 1995 Although documented as having been made in April, this

Sep 20, 1995 change was not actually made in TRNDDB2S until September.

Thanks to Diane Eppestine, Southwestern Bell, USA.
Change 13.030 Support for IBM's IRRDBU00 - RACF Database Unload Utility

EXRAC... is provided by this significant user contribution from

IMACRACF Peter Proppe. The utility creates a sequential file from

TYPERACF a restructured RACF database, so your RACF administrator

VMACRACF can ask complex queries and create reports about the RACF

Apr 18, 1995 security definitions (like who has access to what, etc.!)

There are 39 MXG datasets created by this enhancement.

See comments in member VMACRACF for documentation, and

the IBM documentation is in RACF Macros and Interfaces.

Thanks to Peter Proppe, Bremer Lagerhaus Gesellschaft, GERMANY.

Change 13.029 Landmark MONIDSA build was copied from TYPETMON in Change

TYPEMON8 12.315, but the LABELs and FORMATs for the variables were

Apr 18, 1995 not copied until now. In addition, this "interval" data

record contains neither a DURATM nor STARTIME nor SYSTEM

fields that describe the interval. Am awaiting Landmark.

Thanks to Carl Sommer, SAS Institute Cary, USA.

Change 13.028 Support for Sterling's SOLVE NCL CPU-time accounting user

EXTYSOLV SMF record creates new dataset TYPESOLV allowing you to

IMACSOLV collect NCL CPU usage at the user level. These records

TYPESOLV are interval records, but records are only created if the

VMACSOLV CPU time used by the user exceeds the threshold specified

Apr 14, 1995 in the SOLVE USERACCT command.

Thanks to Arzina Merali, Alberta Public Works, CANADA.
Change 13.027 0C4 ABEND in this assembly program occurred because IBM

ASMRMFV loaded an offset address for enqueue table, but there was

Apr 14, 1995 no enqueue table. To correct, find the line reading

L R8,ENTENTMX(,R9)

and insert these two new lines:

LTR R8,R8

BZ ENDENT

Then find the label (in column 1) MVEENT, and then find

the statement BNE MVEENT several lines later. Insert

(in column 1) ENDENT DS 0H after that BNE.

The actual change in this version of the RMF Monitor III

decompress program was more extensive - now statistics

of records read/written are printed on the job log, and

you can now select which classes of records are written.

Thanks to Leon Berthou, Dan Computer Management, DENMARK.
Change 13.026 ICEBERG SMF record subtype 5 (deleted data space release)

VMACICE did not use ESTOSET to locate the data segment, causing

Apr 13, 1995 incorrect values for the extent addresses and TOIOTIME.

Jun 3, 1995 After the line now reading: EXTOSET=EXTOSET-3;

insert a new line reading: INPUT @EXTOSET @;

Text added 3Jun95:

And delete the line +4 /* UNDOCUMENTED*/

It turns out there have been two different lengths of

ICEBERG (or IXFP) subtype 5 records, one with that 4-byte

undocumented field (and RDW=148), now without that field

(and RDW=144, which means the SAS LENGTH=140). An INPUT

STATEMENT EXCEEDED RECORD LENGTH error occurred when the

140-byte record was read with the 144-byte-expecting MXG

12.12! The above change using EXTOSET and deletion of

the +4 protects for either length of IXFP record.

Thanks to Tom Kelman, SunTrust Service Corporation, USA.

Thanks to Rich Morris, Progressive, USA.

Thanks to Brian Crow, British Telecom, WALES.

Change 13.025 Dataset TYPE72GO variable PGPEXCP is now kept for goal

VMAC7072 mode. Note that PGPIOTM is no longer calculatable for

Apr 12, 1995 Goal Mode, because the IBM developers eliminated the

option to use IOTM instead of EXCP for RMF I/O units in

MVS/ESA 5.1 Goal Mode.

Thanks to Al Keeble, British Steel, ENGLAND.

Change 13.024 The +1 reserved field after the INPUT of CHIRECNT should

VMACTMVS be +4 instead. This error trashed values in the TMVSCH

Apr 12, 1995 (Channel Path I/O) dataset.

Thanks to Nico Lenaerts, SAS Belgium, BELGIUM.

Thanks to Wim Desmecht, DOLMEN Computer Applications NV/SA, BELGIUM.
Change 13.023 Dataset CICFCR variable A17DSTYP was mislabeled, input as

FORMATS numeric instead of character, and was not decoded. Now,

VMAC110 it is input as character, and new format $MGCICDT decodes

Apr 12, 1995 the Dataset Type as BSAM, ESDS, KSDS, RRDS, or PATH.

Thanks to Chuck Hopf, MBNA, USA.
Change 13.022 Support for NETSPY Release 4.6 added, thanks to advance

EXNSPYAL documentation from LEGENT (that release won't ship until

EXNSPYVT summer). Their changes were compatibly made.

FORMATS -Variable NSPYNETW (Network ID) was added to the header so

IMACNSPY it is now kept in every MXG NSPYxxxx dataset.

VMACNSPY -Dataset NSPYAPPL. New variable NSPNETAP (Network ID of

Apr 12, 1995 the Application) was added. Also, labels for variables

Apr 14, 1995 APOTLN62 and APOTNN62 were corrected; these variables

measure bytes sent outbound on sessions where response

time is or is not monitored.

-Dataset NSPYLU. New variables LUNETID, PCSESSID, VILUNAME

were added, containing Network ID of this LU, Procedure

Correlation ID, and Virtual LU Name respectively.

-Dataset NSPYETHR. New variable NSPEDISD, Discarded

Datagrams was added.

-Additional changes were made that were not 4.6 related:

-New datasets NSPYALRT and NSPYVTAM are now created from

type X and B records; these records are not new with

Release 4.6, but had not been decoded previously. The

type X record has been tested with data, but there are

data values for NSPXRTYP and NSPXVART which are not shown

in NETSPY documentation. NSPYVTAM has not been data-ed!

-Corrections found during early code validation:

Test for LMI Station should have been OR instead of AND

i.e., (NSPNSUBE='20'X OR NSPNSUBE='10'X) THEN DO;

Test for FRSE Station should have been OR vice AND:

i.e., (NSPNSUBE='20'X OR NSPNSUBE='80'X) THEN DO;

Second input of REBYTCT with /*XRBY*/ was deleted.

Second REBYTCT in KEEP list for NSPYTIC3 was deleted.

Thanks to Rod Fernandes, Albert Heijn B.V., HOLLAND.

Change 13.021 NETSPY type N records subtypes 07 (TIC3) and 08 (Trans

EXNSPYTP Priority) were not correctly processed, causing Divide by

EXNSPYT3 zero with subtype 07 variable CYCLSPD. The circumvention

IMACNSPY is to change the line

VMACNSPY IF NSPNRECI='02'X OR NSPNRECI='06'X OR NSPNRECI='07' OR

Apr 10, 1995 NCPELTYP=80X THEN DO;

to read

IF NSPNRECI='02' OR NCPELTYP=80X THEN DO;

The actual change was more extensive, creating two new

datasets, NSPYTIC3 and NSPYTPRI from the 07/08 subtypes.

Thanks to Rod Fernandes, Albert Heijn B.V., HOLLAND.
Change 13.020 -XMXGSUM failed if one of the input datasets did not exist

XMXGSUM (as in the first time a TRND members is run) AND any of

Apr 10, 1995 variables in the datasets did not exist. (VMXGSUM did not

fail because it protected by creating missing variables

using the IF X=. THEN X=.; "compiler faker" statement.)

Logic corrected, and switches were added to help in the

debugging of XMXGSUM.

-KEEPIN macro needed to be UPCASE'dED.

Thanks to Diane Eppestine, Southwestern Bell, USA.
Change 13.019 The example SAS program has a "EXPECTING VARIABLE" error.

ACHAP35 -Insert

Apr 3, 1995 TODAY=TODAY();FORMAT TODAY DATE9.; DROP TODAY;

after the statement

IF LASTDATE NE TODAY()-1 THEN DO;

-Change LASTDATE's FORMAT from MMDDYY8. to DATE9. (to

be consistent with MXG's DATE conventions).

-Change "5X" to "+5".

Thanks to Bruce Hewson, Alcoa, AUSTRALIA.
Change 13.018 The %INCLUDE SOURCLIB(ASUMDB2S); should be deleted from

JCLPDB6 the JCLPDB6 example, as there is no member ASUMDB2S.

Apr 3, 1995 (There is a TRNDDB2S, but it uses the DB2STATS dataset

as its input, and DB2STATS is already an interval dataset

so there is no need for an ASUMDB2S. All ASUMxxxx members

create interval data from detail data.)

Thanks to Bruce Hewson, Alcoa, AUSTRALIA.
Change 13.017 TYPE6 OUTDEVCE for a 4-digit Remote Number is R2030PR1,

VMAC6 instead of the old Rnnn.PRm format. MXG parsed variable

BUILDPDB OUTDEVCE to set DEVICE=PRINTER or PUNCH, expecting the

BUILDPD3 period delimiter, but the new format causes DEVICE to be

BUILD005 blank, which causes lines that were formerly counted in

Apr 3, 1995 variable PRINTLNE or PUNCHCRD to be counted in variable

EXTWTRLN instead, in both the TYPE6 dataset and in the

PDB.PRINT and PDB.JOBS datasets.

In VMAC6, replace these four lines:

IF PR=:'PR' THEN DEVICE='PRINTER';

IF PR=:'PU' THEN DEVICE='PUNCH ';

with these six lines:

IF PR=:'PR' OR

(SUBSTR(OUTDEVCE,1,1)='R' AND SUBSTR(OUTDEVCE,6,2)='PR')

THEN DEVICE='PRINTER';

ELSE IF PR=:'PU' OR

(SUBSTR(OUTDEVCE,1,1)='R' AND SUBSTR(OUTDEVCE,6,2)='PU')

THEN DEVICE='PUNCH ';

Note: above text corrected on May 3, from 5,2 to 6,2.
-Variable TOTLINES, the sum of PRINTLNE,PUNCHCRD,EXTWTRLN,

in the PDB.PRINT dataset was always correct. However, I

now realize that TOTLINES should also be created in the

PDB.JOBS dataset, so the BUILDxxx members were changed to

create and label variable TOTLINES into PDB.JOBS.

Thanks to Norbert Korsche, OMV, AUSTRIA.

Change 13.016 MXG 12.12 inadvertently changed some variables' FORMAT

VMAC7072 from 5.1 to 5.0. Line 102100 in VMAC7072 should be 5.1

Apr 3, 1995 instead of 5. so that one decimal place of precision

will be printed.

Thanks to Joanne Turpie, Department of Labour, NEW ZEALAND.
Change 13.015 MXG 12.12 and MXG 12.12A. MONTHBLD sort error; the BY

MONTHBLD list for TYPE72 should be exchanged with the by list for

Apr 3, 1995 TYPE72DL, and in the TYPE72DL sort list, replace PERFGRP

with SRVCLASS. (Make MONTHBLD look like WEEKBLD!).

Thanks to John Mattson, National Medical Enterprises, USA.
Change 13.014 Semicolon is missing at the end of SET PDB.APAFDOMA and

ANALAPAF the colon after PROC PRINTTO DATA=... should be a semi

Mar 30, 1995 colon, and a PROC PRINTTO; was added as the last line to

Apr 25, 1995 reset the print destination back to the log.

Thanks to Bruce Widlund, Merrill Consultants, USA.
Change 13.013 Tracking IMS DEDB (Data Entry Data Base) and MSDB (Main

VMACIMSA Storage Data Base) activity is accomplished with the IMS

Mar 30, 1995 fastpath type 59x subcode 37x log record, but MXG only

output when SYNCRTYP='01'X. In IMS 4.1 at least, most of

the 59x37x records contain either 10x or 20x. Since I

don't normally delete observations, and because the 4.1

documentation suggests the meaning of this field has been

changed, I have changed the logic so all subcode 37 log

records will be output in IMS5937 dataset.

Thanks to Don Cleveland, Blue Cross Blue Shield of California, USA.

Change 13.012 MXG now detects that fields have been EXCLUDED from your

VMAC110 SMF type 110 CICS/ESA records, and prints error message

Mar 30, 1995 telling you to run UTILCICS and use its output to update

member IMACEXCL for MXG to support records with EXCLUDE.

Thanks to George Scott, Rockwell International Corporation, USA.
=====Changes thru 13.011 were in Early MXG 13.01 dated Apr 1, 1995=====
Change 13.011 Support for OS/400 Version 3.1.0 has INVALID DATA FOR ...

VMACQAPM because I had no test data for validation until now!

Mar 29, 1995 -Value of ASLEVEL starts with an unwanted blank. Use the

COMPRESS() function to eliminate that blank, by changing:

AS400VRN=PUT(GDESRV,2.)!!'.'!!PUT(GDESRR,3.1);

to read

AS400VRN=COMPRESS(PUT(GDESRV,2.)!!'.'!!PUT(GDESRR,3.1));

(because ASLEVEL is set from AS400VRN).

-All tests of IF ASLEVEL ... for value of '3' must be

changed to '3.1.0' and the one test for LE '2' must be

changed to LT '3.1.0'.

-Variables SWRES1,SPRES1,SMRES1,S6TRNT,SERES1,SARES1,

SBRES1,SXRES1, and SIRES1 input must be &PD.8 vice &PD.6.

-Variables SIRES1 and SIRES2 are now SITRNT and SITRNS.

-Calculation of PCTIOPBY needs to be copied into the 3.1.0

logic.

-Replace ELSE IF LENGTH=2364 OR LENGTH GE 2374 THEN DO;

with ELSE IF ASLEVEL LE '2.2.0' THEN DO;

-Replace ELSE IF LENGTH=3090 THEN DO;

with ELSE IF ASLEVEL GE '3.1.0' THEN DO;

-Reverse DSTYPE and DSARM in the 3.1.0 input for QAPMDISK.

Thanks to Len Marchant, Coca-Cola Enterprises, USA.

Change 13.010 Complete revision of support for HP's PCS for UNIX. There

EXHPAI.. are now separate members for the AIX, SUN, and HPUX UNIX

EXHPSU.. systems, because not all metrics exist on all three UNIX,

EXHPUX.. and the HPPCS record formats are different on each:

IMACHPAI

IMACHPSU Suffix For Members Infile

IMACHPUX HPAI AIX ADOCHPAI,TYPEHPAI,ASUMHPAI... HPAI

TYPEHPAI HPSU SUN ADOCHPSU,TYPEHPSU,ASUMHPSU... HPSU

TYPEHPSU HPUX HPUX ADOCHPUX,TYPEHPUX,ASUMHPUX... HPUX

TYPEHPUX

VMACHPAI These datasets are created from HPPCS records:

VMACHPSU

VMACHPUX AIX : HPAIAPPL,HPAICONF,HPAIDSK ,HPAIGLOB,HPAILANS

Mar 29, 1995 HPAIPROC

SUN : HPSUAPPL,HPSUCONF,HPSUDSK ,HPSUGLOB,HPSULANS

HPSUPROC

HPUX: HPUXAPPL,HPUXCONF,HPUXDISK,HPUXDSK ,HPUXGLOB

HPUXLANS,HPUXPROC,HPUXVOLS

This change replaces the support in TYPEHPCS.

The HPAI and HPUX support has now been validated in real

production sites. I am still trying to understand the

HP UX Disk data (HPUXDISK and HPUXDSK are similar but not

the same, and have repeated Disk addresses in interval

data) and SUMMARY=5 caused a problem with the AIX data,

but both sites are happy and productivly measuring UNIX!

Thanks to Richard Clary, Entergy Services, Inc.

Thanks to Lorenzo Wright, NCCI, USA.
Change 13.009 DB2 dataset T102S145 variables QW1145OB-QWF145OB have the

VMAC102 DB value instead of the OB value. Find all occurrences

Mar 27, 1995 of QWX145DB, and change the DB to OB in the fifteen lines

with QWn145OB on the left of the equality.

Thanks to Peter Durrant, National Australia Bank, AUSTRALIA.
Change 13.008 Support for TCP/IP SMF record APAR PN69321-PN69322, which

VMACTCP adds the API user's job name, APIJOBNM, to the API INIT

Mar 27, 1995 and API TERM event records. This support was coded based

on the APAR description, but has not been data-tested.

Change 13.007 NetSPY dataset NSPYNCP variables FBQ_PCT, FBH_PCT, and

VMACNSPY FBL_PCT can cause zero-divide error because MXG did not

Mar 24, 1995 understand the new data added by LEGENT.

The block of code that read:

CYCLUTIL=10*UCYCLETM/(DURATM*CYCLSPD);

FBQ_PCT =100*FREEBUFQ/MAXFREEB;

FBH_PCT =100*FREEBUFH/MAXFREEB;

FBL_PCT =100*FREEBUFL/MAXFREEB;

BUF_UTIL=100 - FBL_PCT;

now reads:

CYCLUTIL=10*UCYCLETM/(DURATM*CYCLSPD);

IF TFQBANCP GT 0 THEN DO;

FBQ_PCT =100*CFBQUELN/TFQBANCP;

FBH_PCT =100*FBQLNHWM/TFQBANCP;

FBL_PCT =100*FBQLNLWM/TFQBANCP;