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

when INPUT in the VMACIMS and VMACIMSA members, now that

GMTOFFTM exists in all active IMS Versions; this makes

the detail IMSxx datasets consistent with the final

IMS0708 and IMSTRAN output datasets. The GMTOFFTM logic

in the TYPEIMSA member was removed.

-These five variables are now correctly input as Packed:

DLRMINT /* WAIT TIME*FOR*INTENT*CONFLICT*/

DLRMPOL /* WAIT TIME*FOR*POOL*SPACE*/

DLRMSCH /* ELAPSED TIME*FOR*SCHED PROCESS*/

DLRTMEIO /* DBCTL*DB I/O*ELAPSED*TIME*/

DLRTMEPL /* DBCTL*LOCKING*ELAPSED*TIME*/

-Additional cleanup and validation with V8, V9, and V10

log records, and detection of incorrect _IMSVERS setting

in many cases will stop processing with log messages.

Thanks to Steve Hudson, AVNET, USA.
Change 26.025 Support for APAR OA12774 new z10 variables (COMPAT).

VMAC7072 MXG 25.25 supports (TOLERATE) the new z10 processor, and

VMAC71 it won't fail if you install this APAR, but you will need

Mar 2, 2008 MXG 26.01 if you want to use any of these new variables:

Mar 6, 2008 New variables in TYPE70 dataset:

SMF70HDM='HIPERDISPATCH*IS*ACTIVE?'

SMF70MCR='MODEL*CAPACITY*RATING OF*SMF70MDL'

SMF70MPC='MODEL*PERMANENT*CAPACITY*IDENT'

SMF70MPR='MODEL*PERMANENT*CAPACITY*RATING'

SMF70MTC='MODEL*TEMPORARY*CAPACITY*IDENT'

SMF70MTR='MODEL*TEMPORARY*CAPACITY*RATING'

CPUPATm0='CPU 0*PARKED*DURATION'

CPUPATm1='CPU 1*PARKED*DURATION'

CPUPAT.. ...

CPUPATYb='CPU 62*PARKED*DURATION'

CPUPATYC='CPU 63*PARKED*DURATION'

CPUPATTM='TOTAL*PARKED*DURATION*ALL CPUS'

New variables in TYPE70PR dataset:

SMF70CAN='AVERAGE*THIS SMF70CIN*ENGINES*IN CEC'

SMF70POW='AVERAGE*HIPERDISPATCH*WEIGHT'

SMF70PAT='LCPUADDR*PARKED*DURATION'

New variables in TYPE70Y2 Crypto dataset:

R702NH5C='SHA-384/512:CALLS*TO*HASH'

R702NH5B='SHA-384/512:DATA*BYTES*HASH'

R702NH5I='SHA-384/512:PCMF INSTR*USED TO*HASH'

CRYIH5R ='SHA-384/512*HASHING*RATE'

CRYIH5S ='SHA-384/512*HASHING*SIZE'

New variables in TYPE71 dataset:

SMF71LOA='AVG*LARGE MEMORY*OBJECTS*ALLOCATED'

SMF71LOM='MIN*LARGE MEMORY*OBJECTS*ALLOCATED'

SMF71LOX='MAX*LARGE MEMORY*OBJECTS*ALLOCATED'

SMF71LRA='AVG NR*1 MB FRAMES*BACKED*REAL STORE'

SMF71LRM='MIN NR*1 MB FRAMES*BACKED*REAL STORE'

SMF71LRX='MAX NR*1 MB FRAMES*BACKED*REAL STORE'

-Discovered: these 12 percentage vars for CPUIDs 10,11,12

PCTCPBYA-YC, PCTIFBYA-YC, PCTZIBYA-YC, PCTCIBYA-YC

were overlaid by values from CPUIDs 61,62,63 if they

existed, because those names were reused.

New variable names for these metrics are now:

PCTCPBXA-XC, PCTIFBXA-XC, PCTZIBXA-XC, PCTCIBXA-XC

-Discovered: these variables summed across all CPUIDs

did not include CPUIDs 54 thru 63 (suffix ZT-ZZ,YA-YC):

CPUWAITM MVSWAITM CPUPDTTM CPUPATTM CPUEFFTM
Change 26.024 DB2STATS variables QTPACAUT, QTPACNOT and QTPACPUB were

VMACDB2 not de-accumulated.

Mar 1, 2008
Change 26.023 MXG 25.07-MXG 25.25 only. Fewer obs created in TYPE78CU

VMAC78 and in TYPE72IO datasets. The last LCUID was never output

Feb 27, 2008 in TYPE78CU because MXG's test for 60 bytes remaining

Apr 11, 2008 should have been for 58 bytes after the pointer was at

the end of the 2-byte field.

Thanks to Steve Lottich, The University of Iowa Hospitals, USA.

Change 26.022 The RACF Unload utility IRRDBU00 was revised to support

VMACRACF reading the EBCDIC text file while MXG executes on ASCII.

Feb 26, 2008 The output file is all text/ZD fields and MXG assumed

that under ASCII the data would have been downloaded and

translated from EBCDIC, but using ftp access to read the

file from ASCII did not work without revisions.

-Also, there were '92'x ASCII text (instead of '27'x for

single quotes) that when uploaded to z/OS became '00'x,

a non-printable, fortunately inside comments in VMACRACF

so they had no impact. But then a search of all members

found those text were also in the VMAC122, NEWSLTRS, and

DOCVER members, which were also corrected to '27'x text.

Thanks to Paul Gillis, Pacific Systems Management Pty., AUSTRALIA
Change 26.021 Specifying OUTDATA=CICSVIEW and OUTVIEW=YES with VMXGSUM

VMXGSUM caused errors because there was an extraneous &VWUTDATA

Feb 26, 2008 reference outside the DO group in line 1874. Deleting

the line corrects the MXG error.

Thanks to Nigel Greenwood, EDS, UK.
Change 26.020 Support for optional user-created CICS fields:

IMACICU6 CMODHEAD CMODNAME Variables Created

IMACICU7 VZSOAP SOAPOUT SOAPOUCN, SOAPOUTM

UTILEXCL VXAOAP SOAPABCT SOAPABCT

VMAC110

Feb 24, 2008

Thanks to Herbert Sweeney, Verizon Data Services Inc, USA.

Thanks to Vernon W. Gordon - Verizon Data Services Inc, USA.

Thanks to Marshall S. Hellmann - Verizon Data Services Inc, USA.

Thanks to Michael F. Root - Verizon Data Services Inc, USA.

Change 26.019 The DURATM values in IMACSHFT are used only for _DURSET,

IMACSHFT but the values in the examples were wrong. For those

Feb 20, 2008 three example shift definitions, the DURATM values should

have been 9*3600, 15*3600, and 48*3600

Thanks to Jeffrey A. Harder, Indiana Farm Bureau Insurance, USA.
Change 26.018 Dataset BVIR30 contained only data for Preference Group

VMACBVIR Zero, PG0. These PG1 variables are now INPUT and KEPT:

Feb 20, 2008 VIRVOL01 DATRES01 AV4HAG01 VL4MIG01 AV48AG01 VL48IG01

AV35AG01 VL35IG01

Thanks to Stan Adriaensen, AXA, BELGIUM.
Change 26.017 The COMPBYTE and UNCOMBYT byte counts were increased to 8

VMACXCOM bytes with two overflow fields that are input and added.

Feb 19, 2008 One new field overlaid TCPIPPNR, so it's a missing value.

Thanks to Peter J. Gray, EDS, AUSTRALIA.

Change 26.016 BLDSMPDB options WEEKKEEP WEEKDROP MNTHKEEP MNTHSROP did

BLDSMPDB not protect for lower case in the user input text.

Feb 18, 2008

Thanks to Loren Mitchell, Los Angeles County Office of Education,USA.

Change 26.015 Variable R783HLCU was not kept nor labeled, even though

VMAC78 I used it to confirm SMF78HIX was wrong, which led to

Feb 18, 2008 IBM APAR OA21799 (see Change 25.141)!

Thanks to Christa Neven, KBC Bankverzekerinngsholding, BELGIUM.

Change 26.014 IMPLX Version 4.1 support was incorrect; six fields that

VMACMPLX were input as PIB4. are actually PIB8., causing trashed

Feb 15, 2008 values in many variables.

Thanks to David Bixler, FISERV, USA.

Change 26.013 -With PARM='NONE', ASMRMFV still attempted to open the

ASMRMFV DDNAMES RMFBSAM, RMFFILT, and RMFSKIP. NONE was intended

Feb 10, 2008 to product no output, but omitting //RMFBSAM resulted in

a spurious USER ABEND, now eliminated. The //RMFBSAM DD

is still required for all other RMF table based selection

-Comments documenting RMFBSAM and "NONE" were enhanced.

-PARM='NONE' caused unnecessary ASMRMFV processing: the

output records were completely built before the test for

NONE, which prevented the write of the record. The test

was relocated to bypass that unwanted processing.

Thanks to Jack Schudel, University of Florida, USA.
Change 26.012 Cosmetic, used by MXG technical support. VMXGINIT prints

VMXGINIT the z/OS full dataset name or the ascii full path name

Feb 8, 2008 of each of the SOURCLIB and SASAUTOS concatenations.

On ASCII, a FILENAME statement with the LIST; operand is

used, but that does not work on z/OS, so a DATA step that

reads VEXTFL is used:

DATA; SET SASHELP.VEXTFL (UPCASE(FILEREF)='SOURCLIB'));
Change 26.011 Change 25.306 did NOT support IFCID 22 APAR PK38803; that

EX102I22 change text claimed SMF records with the APAR were wrong,

IMAC102 but that was my error; the SMF records with the APAR do

VMAC102 match IBM's DSECT, which I misread. The T102S022 dataset

VMXGINIT created by Change 25.306 will only have valid data in the

Feb 8, 2008 first obs output from each record; subsequent segments

were mis-aligned and had trashed values.

Moreover, MXG output T102S022 for each INDEX in each plan

keeping repeated values for the plan fields and wasting

disk space. This change creates new T102I022 Index

dataset with an obs for each index, keeping only the

index and sort variables, and creates the existing

T102022 dataset with one obs for each plan segment,

keeping only the plan related variables.

These index variables were removed from T102S022 dataset

QW0022MS QW0022NM QW0022AC QW0022AN

QW0022MO QW0022XF QW0022FF QWOL22AC

QW0022AC QWOL22AN QW0022AN

as they are now only kept in the T102I022 index dataset.

Thanks to Scott Barry, SBBWorks, Inc, USA.

Change 26.010 New utility %MACRO VMXGPARS parses macro variables, like

UTILBLDP MACKEEP=, INCODE=, etc., that are a long string of text,

VMXGPARS into "chunks" that fit in 72 bytes and that are chunked

Feb 8, 2008 into reasonably humanly readable pieces of code. Used

now in UTILBLDP, it's use will like be extended to other

%MACRO defining members in the future.

Thanks to Chuck Hopf, Bank of America, USA.
Change 26.009 NOTE: INVALID DATA FOR TRNSTTRA and TRNSTOPA is cosmetic

VMACCIMS only, and occurs when either field is not populated. MXG

Feb 6, 2008 now protects the PD input with double question marks.

The output datasets were correctly created.

Thanks to Ermanno Bertolotti, Intesa SanPaolo, ITALY.
Change 26.008 New MYDATA parameter is added, and a new CASE 3 example

VMXGUOW of how you might use it to insert new dataset name(s) and

Feb 5, 2008 a keep list, so that you can create a subset of ASUMUOW

(eg, based on Transaction name), without having to repass

millions of records in ASUMUOW to create it.

Thanks to Chuck Hopf, Bank of America, USA.

Change 26.007 The CICS Dispatcher Statistics CICDS dataset created from

VMAC110 SMF 110 Subtype STID=60 statistics segments, and the MXG

VMXGCICI PDB.CICINTRV interval summary dataset have new variables

Feb 5, 2008 CICTCTTM='TOTAL*DSGTCT*DURATION*TIME'

CICTDTTM='TOTAL*DSGTDT*DISPATCHED*TIME'

CICTWTTM='TOTAL*DSGTWT*IN OS WAIT*TIME'

and the existing

CICTCBTM='TOTAL*DSGACT*CPU TCB*TIME'

summed across all 19 CICS TCB names for the interval.

These data were added primarily so an IBM "Rule of Thumb"

could be tested; new variable

PCTREGBY='PCT*TCB/(TDT+TCW)*BUSY;

calculates the percent of Dispatch+OS Wait time that was

executing CPU TCB time, and IBM's recommends that if the

PCTREGBY formula calculates to more than about 70 percent

busy, they suggest that you need additional CICS Regions.

Thanks to Jacob Nudel, Thompson BETA Systems, USA.
Change 26.006 Support for 64 CP Engines adds sets of variables for CPs

VMAC7072 54 thru 63 with suffixes ZS-ZZ and YA-YC for TYPE70 data.

Feb 1, 2008 No such systems yet exist, but are expected so the code

Mar 8, 2008 is now in place to create those per-CPU variables.

Change 26.005 In all XAM VM datasets, variable ZDATE was LENGTH 5,

VMACXAM because it was created by code in VMACXAM instead of the

Jan 31, 2008 %%INCLUDE SOURCLIB(IMACZDAT);

which sets ZDATE's length to 4, which is always used to

create ZDATE, not only for length consistency, but the

IMACZDAT member is the tailoring member you would use for

a rerun of a prior daily build PDB job, to set the value

of ZDATE to that prior date. Having LENGTH 5 is not an

error, even though only 4 bytes are used elsewhere in MXG

but it's inconsistency uncovered this inconsistency.

Thanks to Theo Peelen, Fortis Bank, BELGIUM.
Change 26.004 -MXG expected NMON's "AAA,time" record ahead of "AAA,date"

VMACNMON but this site's data had "AAA,date", several lines, then

Jan 31, 2008 "AAA,time", which caused STARTIME to always be a missing

value. Presuming a local specification changed the order

I protected STARTIME for either order, found STARTIME was

now always populated in NMONINTV, but was still always

missing in NMONCPUD and all of the other datasets. After

much debugging, I discovered the original NMON data file

had been SORTED, which not only put date ahead of time,

but had lumped all of the end of interval 'ZZZZ' records

at the bottom, so none of the ENDTIME-only records, like

the CPUnn records, had either ENDTIME or STARTIME.

The moral: DO NOT SORT NMON DATA;

-Support added for new record types: ESSREAD and ESSWRITE

with seven VPATHn KB/sec, and ESSXFER with seven transfer

rates per second for each vpath.

Thanks to Massimo Pugliese, ELSAG Spa, ITALY.
Change 26.003 LPARCPUS is not integer in PDB.ASUM70LP & PDB.ASUMCELP

VMXG70PR MXG summary datasets, and LPARCPUS=0 could have been a

Feb 5, 2008 value between 0 and 1, because MXG erroneously FLOOR'd

LPARCPUS to force an integer value. While LPARCPUS in

PDB.TYPE70PR observations at the LCPUADDR detail is an

integer number of CP engines, in summarizing counts of

LCPUADDRs in each LPAR, and/or multiple intervals, MXG

sums SMF70ONT (the Online Time) and divides by DURATM,

the interval duration to create LPARCPUS. But the IPL

interval for a 3-CP engine with DURATM 5 minutes had only

3 minutes of uptime, SMF70ONT, causing LPARCPUS=0, when

the true LPARCPUS=0.7 value should have been calculated.

Now, that FLOOR is removed, LPARCPUS is FORMATed 6.1,

(so the printed value will be rounded and hence those old

2.9999 values will print as 3.0), and is now labeled:

LPARCPUS='LPAR*AVERAGE*NUMBER OF*CP*PROCESSORS'

in the datasets built by ASUM70PR summarization program.

Thanks to William Wai Lun WONG, HSBC, HONG KONG.

Change 26.002 Cosmetic, but still wrong and possibly confusing: a dozen

FORMATS value statements in FORMATS had incorrect right-hand hex

Jan 31, 2008 for the left-hand value (like '04'X='40X: explanation').

A new QA job will catch this type of error in the future.

Thanks to Danny Ball, Con-way Enterprise Services, USA.
Change 26.001 Typo VMUM introduced in Change 25.252 corrected to VWUM.

VMXGCAPT

Jan 30, 2008

Thanks to Brian Carter, EDS, UK.

LASTCHANGE: Version 26.

=========================member=CHANGE25================================

/* COPYRIGHT (C) 1984-2008 MERRILL CONSULTANTS DALLAS TEXAS USA */

MXG Version 25.25 is dated Jan 28, 2008, thru Change 25.309

First MXG Version 25.25 was dated Jan 21, 2008, thru Change 25.302

MXG Version 25.12 was dated Jan 21, 2008, thru Change 25.294

MXG Version 25.11 was dated Dec 7, 2007, thru Change 25.268

First MXG Version 25.11 was dated Nov 22, 2007, thru Change 25.256

MXG Version 25.10 was dated Oct 7, 2007, thru Change 25.215

MXG Version 25.09 was dated Sep 17, 2007, thru Change 25.199

MXG Version 25.08 was dated Sep 5, 2007, thru Change 25.187

First MXG Version 25.08 was dated Sep 4, 2007, thru Change 25.182

MXG Version 25.07 was dated Aug 10, 2007, thru Change 25.161

MXG Version 25.06 was dated Jul 4, 2007, thru Change 25.134

MXG Version 25.05 was dated Jun 7, 2007, thru Change 25.107

MXG Version 25.04 was dated May 7, 2007, thru Change 25.086

Second MXG Version 25.03 was dated Apr 12, 2007, thru Change 25.058

First MXG Version 25.03 was dated Apr 10, 2007, thru Change 25.057

MXG Version 25.02 was dated Mar 10, 2007, thru Change 25.033

MXG Version 25.01 was dated Mar 5, 2007, thru Change 25.025

MXG Version 24.24 was dated Feb 5, 2007, thru Change 24.306

MXG Newsletter FORTY-NINE was dated Feb 5, 2007

MXG 25.25 is the 2008 "Annual Version", dated January 28, 2008.
Instructions for ftp download of MXG 25.25 were mailed to all licensees,

but you can always request ftp download instructions using the form at

http://www.mxg.com/ship_current_version
Contents of member CHANGES:
Member NEWSLTRS (and the Newsletters frame at http://www.mxg.com) now

contain the current MXG Technical Notes that used to be put in member

CHANGES between Newsletters. New Technical Notes are now added (and

now dated!) in NEWSLTRS/Newsletters with each new MXG Version.

I. Current MXG Software Version 25.25 is available upon request.

II. Incompatibilities and Installation of MXG 25.25.

III. Online Documentation of MXG Software.

IV. Changes Log

=======================================================================

I. MXG Version 25.25, now re-dated Jan 28, 2008.

Minor cosmetic corrections in MXG 25.25, re-dated Jan 28, 2008
TYPEXAM 25.307 Extraneous PUT from testing removed.

ASUM70PR 25.303 Extraneous PROC PRINT from testing was removed.

Minor enhancements in MXG 25.25, re-dated Jan 28, 2008
TYPERMFV 25.309 Support for RMF III CPD, Channel Path Data Table.

TYPE102 25.306 IBM SMF 102 IFCID 22 APAR PK38803 wrong, supported.

TYPE30 25.304 Support for APAR OA23679, BLKSIZE might be wrong.

COMPINTV 25.302 Capture/Compare ALL CPU times in 30/70/72/100/101/110

Major enhancements added in MXG 25.25, dated Jan 21, 2008
ADOCs 25.296 All ADOCxxxx member's CONTENTS were updated.
Major enhancements added in MXG 25.12, dated Jan 21, 2008
TYPEDB2 25.265 DB2TCBTM Correction for DB2 V9.1, REQUIRED.

TYPEDB2 25.291 DB2ACCT variable QWACUDCP CPU time added in DB2TCBTM.

TYPE122 25.287 Support for SMF 122, Tivoli Tape Allocation Manager

TYPEITRF 25.286 Support for ITRF DCR77/DCR78 additions.

TYPENTSM 25.282 Support for new NTSMF Objects, Database Mirroring.

TYPEMQLG 25.280 Support for MQ V6 System Admin Accounting Queue Log.

TYPE85 25.279 Support for subtypes 38, 39 and 40 for ID=85 OAM.

TYPEPSYC 25.277 Support for PSYNCH/390 SMF record.

TYPE22 25.276 Support for APAR OA20043 Dynamic Volume Expansion

TYPE102 25.306 Support for APAR PK38803 INCOMPAT SMF 102 IFCID 22.

TYPE50 25.269 Support for OSA-MPC VTAM SMF 50 subtype 4 record.

TYPETIMS 25.271 Revisions for TMON/IMS support

ASUM70PR 25.270 INTERVAL=DURSET cannot be used.

CONFIGV9 25.267 VALIDVARNAME=V7 added to SAS CONFIGs.

ITRMTNG 25.262 233 DDU files to create ITRM definitions for TYPETNG.
Major enhancements added in MXG 25.11 dated Dec 7, 2007.
TYPE111 25.241 Support for CICS CTG 7.1.0 new SMF 111 record.

TYPE110 25.240 Full Support for CICS/TS 3.2 Compressed Data

EXITCICS 25.240 New "CICS" INFILE EXIT for CICS compressed SMF data.

TYPE7072 25.224 CPUTYPE tests are replaced with ZARCHMDE tests.

This means that with MXG 25.11 or later, a new IBM

CPUTYPE will NOT require a new MXG Version.

TYPETPMX 25.239 Support for Thruput Manager SLM and DB2 data.

TYPE82 25.257 Support for ISCF HCR7750 TKE Logging update.

TYPEEVTA 25.255 Support for Action Software's EventAction user SMF.

TYPE85 25.234 New variables in OAM subtype 32-35 records.

TYPE78 25.236 Zero obs in TYPE78IO with Change 24.171 if z/OS 1.7.

TYPEEVTA 25.255 Support for Action Sofware's EventAction SMF record.

TYPERMFV 25.246 Updates for CPU Segmentation changes.

TYPENTSM 25.253 Support for new NTSMF objects for MSSQL.

TYPETNG 25.221 Support for VM Ware VSX Systems in CA NSM records.

TYPETNG 25.235 New Solaris, AIX, and many RedHat objects added.

VMXGSUM 25.248 New &LNSUMOUT=8 will make all output to length 8.

UTILEXCL 25.256 Macro variable &MXGDEBUG revised for IMACEXCL plus!

EXITCICS 25.240 MCTSSCRL now tested vs MCTMNOPN for CICS Compressed.

TYPE110 25.240 MCTSSCRL now tested vs MCTMNOPN for CICS Compressed.

UPRINDOC 25.226 Utility prints NAME and LABEL of all variables.

TYPE30 25.260 MXG 25.10, INTRVLTM missing for TYPETASK='OMVS'

ANALACTP 25.254 Sample report summarizes DB2 Package data to UOW.

CONFIGW2 25.252 MXG updates for testing MXG Execution under WPS.

Major enhancements added in MXG 25.10.
TYPE7072 25.205 Support for z/OS 1.9, up to 54 CPUs per MVS, INCOMPAT

TYPERMFV 25.204 CFI Segmentation eliminates RMF III skipped CF data.

ANALDB2R 25.202 VARIABLE QBnTDPIO NOT FOUND error corrected.

TYPE70 25.211 ZIPACTTM, PCTZIPBY corrected for Dedicated zIIPs.

ASUMCELP 25.209 Duplicate observations in PDB.ASUMCELP eliminated.

TIMEBILD 25.209 Optional SYNC59 timeshifting using TIMETABL.

Major enhancements added in MXG 25.09.
IMPORTANT CHANGES:
Almost none! Only UTILEXCL in 25.08 had an error, but these other

fixes/enhancements are now ready for prime time:

UTILEXCL 25.193 MXG 25.08 ONLY: LABEL IMACICU3 NOT FOUND.

TYPERMFV 25.191 Support for RMF Monitor III CFI table enhancements.

TYPESRDF 25.195 Support for EMC's SRDF/A user SMF record.

READDB2 25.189 New PDBOUT=YES, old PDBOUT= changed, writes to WORK.

ANALDB2R 25.189 New PDBOUT=YES, old PDBOUT= changed, writes to WORK.

RMFINTRV 25.199 SMF70GIE now reset to 00/15 if SYNC59=YES is used.

TYPE89 25.198 SMF89HOF,SMF89DTO were incorrect due to typo.

UTILCSV 25.197 %UTILCSV creates a CSV (or TAB) Delimited flat file.

UTILBLDP 25.196 Large &MACKEEP string caused strange results.

TYPE92 25.192 New ID=92 ST=14 INPUT EXCEEDED if not a RENAME.

Major enhancements added in MXG 25.08.
IMPORTANT CHANGES:
TYPETNG 25.181 Support for CA NSM RedHat 4.01 Linux perf cube.

TYPE7072 25.176 Support for APAR OA18244, Blocked Workload z/OS 1.9.

TYPE7072 25.163 Support for Capacity Groups variables in TYPE70.

ASUM70PR 25.163 Capacity Group summarization, PDB.ASUM70GC/ASUM70GL.

TYPEQACS 25.178 AS400 APAR QAPMDISK with LRECL=456 added new data.

ADOCDB2 25.172 Example to process DB2 datasets to separate DDNAMES.

TYPEDB2 25.169 _RDB2ACC DB2 Parallel event "analysis" example.

Many 25.179 %UPCASE,%LOWCASE,%STR,%BQUOTE,%QUOTE, etc.

Doc 25.179 Use %LET MACxxxx= %STR( text ) ; to pass text.

Major enhancements added in MXG 25.07.

CRITICAL CHANGE:
TYPE78 25.141 APAR OA21799 for HiperPAV, ABEND, SMF78HIX invalid.

Installing HyperPAV can create invalid RMF 78-3's

that cause BUILDPDB to ABEND as it reads RMF 78's.

Change 25.141 will detect bad records and avoid the

ABEND, but you will need to install the IBM PTF for

the APAR to correct the invalid data values.

IMPORTANT CHANGES/ENHANCEMENTS:
Many 25.140 Prelim z/OS 1.9 (fails if 54-CPs, See Ch 25.205)

TYPECIMS 25.139 Support for IMF Version 4.3 (INCOMPATIBLE).

TYPE74 25.140 APAR OA17070 supports CF Level 15 measurements.

TYPE89 25.138 Support for APAR OA20314 new SMF89LPN/SMF89ZNA.

TYPE80A 25.137 Support for unknown TOKDANAMs, prevents ABEND.

UTILLPDS 25.136 Utility to count used/defined PDS Directory Blocks.

TYPE7072 25.135A LCPUCAP/LCPUCAPC Labels include "Hard CAP".

TYPE42 25.153 MXG 25.06 only, false INVALID TYPE42 SUBTYPE 5 error.

TYPEVMXA 25.151 180 Error _MPRCAPC not found, DEBUG prints removed.