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

TYPEIAM 35.107 Support for IAM Version 9.0.

TYPEIMS 35.058 Support for IMS LOG 67D0 DIAGNOSTIC DC Log Record.

TYPEMVCI 35.062 Support for Mainview CICS CMRDETL file VER 6700.

TYPEMVCI 35.161 Support for BMC Mainview/CICS Version 7.1.

TYPEMVIP 35.055 Support for Mainview for IP PTF BPN2331 adds flag.

TYPEMVJE 35.094 Support for BMC Mainview for Java Environment.

TYPENMON 35.208 Nigel's Monitor changed HH:MM to N MINS, INCOMPAT.

TYPEOPAV 35.163 Support for Dell/EMC Mainframe Enabler PAV Optimizer

TYPEOPC 35.048 Support for IWS/TWS/OPC Version 9.3 ST 66 was ST 23.

TYPEOPSS 35.090 Support for CA's OPSS Product User SMF Record.

TYPEOSEM 35.010 OSEM User SMF INPUT EXCEEDED, invalid, circumvented.

TYPEPOEX 35.002 INVALID SMF Records caused STOPOVER ABEND.

TYPEPOEX 35.242 Support for Power Exchange Version 10.1.1.

TYPEPOEX 35.257 Protection for truncated Power Exchange SMF record.

TYPEQACS 35.288 Support for QAPMDISK with LENGTH=695.

TYPERMFV 35.005 Dataset ZRBLCP obs created for ONLINE LCPUADDRs.

TYPERMFV 35.028 New RMF III ZRBENC "long names" now input and kept.

TYPERMFV 35.044 ZRBCP SMT vars missing, new CPC_CECNAME variable.

TYPERMFV 35.191 Support for z/OS 2.3 ZRBASI and ZRBUWD new fields.

TYPERMFV 35.235 RMF III ZRBCPU enhanced with decodes of CPC_HOMEFLAG.

TYPERMFV 35.259 35.10: ZRBASI deaccumulation was not correct.

TYPERMFV 35.259 IBM 4HR MSU (CPUAVB4H) in ZRBCPU per) interval.

TYPERMFV 35.259 MSU Count variables added to ZRBASI/ZRBCPU/ZRPLCP.

TYPERMFV 35.259 New ZRBLCPLPAR dataset with per-LPAR totals.

TYPERMFV 35.287 MXG 35.10/35.11 RMF III ZRBASI ASICPUTA was WRONG.

TYPERMFV 35.300 The CPUPHYAD format could fail creation with ABEND.

TYPEROSC 35.177 PDB.ROSCOE, ROSIGNON Logon Time, CONNECTM, corrected.

TYPESVIE 35.059 Support for CA SYSVIEW for IMS 14, missing values.

TYPETMS5 35.278 Correction for TMS Stacked Tape Files wrong values.

TYPETPMX 35.261 Execution time for TYPETPMX halved by restructure.

TYPETPMX 35.261 Execution time reduction.

TYPETPX 35.035 Protection for invalid TPX subtype 7 record.

TYPETPX 35.155 STOPOVER when IP Port was changed from 4 to 5 digit.

TYPEVMXA 35.025 Using _VMINPUT. z/VM variable VMDUSER was 1 byte.

TYPEVMXA 35.040 Support for Velocity ZWRITE z/VM MONWRITE records.

TYPEVMXA 35.079 z/VM 6.3 SMT in VXSYTPRP, VXAPLSO0 corrections.

TYPEVMXA 35.092 Additional support for z/VM 6.4 (INCOMPAT, SYTLCK).

TYPEVMXA 35.093 MXG 35.03 only. PLATBUSY/PCTOF HDW TYPE70/RMFINTRV.

TYPEVMXA 35.131 Variable CALENMT incorrect, new CALSHARE variable.

TYPEVMXA 35.132 Support for zVM 6.4 APAR VM66026 new variables.

TYPEVMXA 35.145 zVM SMT INTERVAL vars were incorrectly DIF()'d.

TYPEVMXA 35.165 New variables added to VXMTRMEM dataset.

TYPEVMXA 35.174A MONWRITE VXBYUSR _MT1 and _PRO (SMT times) corrected.

TYPEVMXA 35.203 z/VM 6.4.17.1 INCOMPATIBLE fields.

TYPEVMXA 35.221 zVM MONWRITE VXPRCPUP dataset corrected.

TYPEXAM 35.063 Support for XAMSYS wrong length, XMTCPSYS NAMENODE.

TYPEXAM 35.074 Velocity XAM SYTCPU invalid errors at vendor.

TYPEXAM 35.164 New variables added to XAMSYS dataset.

TYPEXAM 35.195 Support for zVPS XAM XAMPUP segment.

TYPEXAM 35.218 XAMSYPUP dataset variables are now correctly aligned.

TYPEXAM 35.223 zVPS/XAM extra SYTCUP with totals is now decoded.

TYPEZDP 35.162 Support for Dell/EMC Mainframe Enabler zDP

UTILBLDP 35.143 Options SUPPRESS enhanced, NEVER corrected.

UTILBLDP 35.225 New EXPDBVAR/EXPDBCDE/EXPDBOUT to create subset.

UTILBLDP 35.306 SUPPRESS=74 variable DEVN NOT FOUND ERROR.

UTILBLDX 35.149 New BUILDJCL=YES uses IFASMFDP to save CPU time.

UTILCMPR 35.292 Utility compares numeric variables in OLD/NEW dataset

UTILEXCL 35.004 MXG 34.34 PDB.CICSDICT not FOUND - GET NEW UTILEXCL.

UTILEXCL 35.023 MXG 35.01.Old Dictionary Records were not used.

UTILEXCL 35.228 Support for 20 user character fields in CICSTRAN.

UTILEXCL 35.293 &MXGCIEXC "exit" to correct USER CMODHEAD typos.

UTILRMFI 35.026 Enhanced reporting if SRVCVLASS=SYSOTHER detected.

VGETSORT 35.112 35.04 only. ERROR Truncated SORTBY (name GT 32).

VGETxxxx 35.309 Protection for DATASET=PDB.dataset syntax.

VMAC38 35.136 NETVIEW ID=38 unexpected S38CCALR length corrected.

VMACSMF 35.266 SMF ID=2 SYSTEM=DUMY 14 byte records protected.

VMXGALOC 35.033 Month begin/logic revised, MNTHKEEP zero protected

VMXGFIND 35.117 Multiple input PDBs were read, only one was output.

VMXGPRNT 35.120 WPS only, MXG 35.04 Only, Blank Label ERROR.

VMXGSET 35.256 Example to read "concatenated" PDBs with PROC SQL.

VMXGSUM 35.020 MXG 35.01. Disregard MXGWARN VMXGSUM BACKLEVEL msg.

VMXGSUM 35.056 Correction for KEEPMNTH= (very rarely used) option.

See member CHANGESS for all changes ever made to MXG Software, or

the CHANGES frames at http://www.mxg.com.

Inverse chronological list of all Changes:
NEXTCHANGE
====== Changes thru 35.309 are in MXG 35.36 dated Jan 13, 2018=========
Change 35.309 All of these macros have both a DDNAME and a DATASET

VGETFMT parameter but if you specified DATASET=PDB.dataset they

VGETLABL would all fail since they looked for WORK.PDB.dataset.

VGETLEN This change looks at the code and if DDNAME is null looks

VGETVAR at dataset and uses the first 'word' delimited by '.' of

Jan 8, 2018 DATASET as the DDNAME and the second for the DATASET. If

there is no '.' then DDNAME is set to &MXGWORK.
Change 35.308 SAS defect in SAS 9.4 M5 PROC SQL, only M5 on z/OS, when

ASUMUOW PROC SQL is executed after OPTION OBS=0 was set, caused

Jan 6, 2018 "SQL SET NOEXEC OPTION" error message and ERRORABEND.

Error occurred in default ASUMUOW, but ONLY if you did

NOT enable IMACUOW to create observations, as then, MXG

sets OBS=0 prior to this failing PROC SQL (which had no

prior error message than the NOEXEC and which is still

under investigation by SAS Support: SAS NOTE 61672.

The circumvention is to add NOERRORSTOP to this PROC SQL

and to the several hundred other PROC SQLs in 51 members,

and do it now to hopefully avoid the need for a SAS fix.

Most of the SAS examples of PROC SQL use NOERRORSTOP and

no MXG written PROC SQL has ever had a syntax error, so

this circumvention will likely be permanent.

Option ERRORSTOP is the SAS Default for batch, and it

determines whether PROC SQL stops executing if it

encounters an error; option NOERRORSTOP instructs PROC

SQL to execute the statements and to continue checking

the syntax after an error occurs.
Change 35.307 ANALDB2R fails with broken DO syntax due to Change 35.263

ANALDB2R (MXG 35.11) which incorrectly set the count of SORTBY=

Jan 6, 2018 arguments, resulting in an error in the data steps, if

option PDBOUT=PDB is used.

Thanks to Harald Seifert, HUK-COBURG, GERMANY.
Change 35.306 UTILBLDP with SUPPRESS=74 variable DEVN NOT FOUND ERROR.

UTILBLDP If you specified SUPPRESS=74 the sort of the TYPE74CA

Jan 6, 2018 dataset failed with BY variable DEVN not found.

When UTILBLDP found RMF datasets that are needed for

RMFINTRV are SUPPRESSED, we chose to still create them in

WORK so that RMFINTRV would find them and not fail.

But a change to SUPPRESS logic nulled MACRO _CDE74 and

only 16 variables were kept (those that were INPUT in the

other _CDEnnnn macros), and the _STY74 failed.

This change reinstates the logic that nulls the _Sxxxx

macro for suppressed RMF records so sorts will not fail,

but if you also want RMFINTRV to be valid, but don't want

the high volume TYPE74's processed, then you should use

ZEROOBS=74 so the datasets will be created but with zero

OBS, so RMFINTRV will be happy.
Change 35.305 -Jan 04: MXG 35.35 didn't protect LENSR=480 length, caused

VMAC42 STOPOVER if you happen to have that length/APAR.

Jan 4, 2018 -Jan 04, IBM confirmed their incorrect values and will now

set SRLEN=160, and note that that does NOT include the

SYNC segment's 80 bytes when present.

-Change 35.302 in MXG 35.36 was the original change.

-Change 36.023 in MXG 36.01 added invalid LENSR=232.
Change 35.304 New variables in TYPE71 in z/OS 2.3:

VMAC71 SMF71L8M ='MIN 1MB*FRAMES*IN CSTORE'

Jan 4, 2018 SMF71L8X ='MAX 1MB*FRAMES*IN CSTORE'

SMF71L8A ='AVG 1MB*FRAMES*IN CSTORE'

SMF71L9M ='MIN 1MB*AVAILABLE*FRAMES*IN CSTORE'

SMF71L9X ='MAX 1MB*AVAILABLE*FRAMES*IN CSTORE'

SMF71L9A ='AVG 1MB*AVAILABLE*FRAMES*IN CSTORE'

SMF71L10M='MIN 1MB*FRAMES*IN-USE BY*MEM OBJECTS'

SMF71L10X='MAX 1MB*FRAMES*IN-USE BY*MEM OBJECTS'

SMF71L10A='AVG 1MB*FRAMES*IN-USE BY*MEM OBJECTS'

====== Changes thru 35.303 are in MXG 35.35 dated Jan 3, 2018=========
Change 35.303 One z/OS SAS 9.4 M5 site gets SQL SET NOEXEC OPTION that

VMXGUOW terminates the job, currently under investigation by SAS

Jan 4, 2018 support, but adding NOERRORSTOP option to the PROC SQL

does circumvent the error, so it has been added to the

one failing PROC SQL in hopes thats the only one needed.

This note will be revised when more is known.

Change 35.302 SMF 42 st 5/6 with OA54112 now has three SRLEN values of

VMAC42 of 240 and 400, and 480 from OA52132/OA52133/OA61745 in

Jan 2, 2018 Change 35.289, and all three are wrong.

Jan 3, 2018 The actual length of the SR segment in each record is

Jan 4, 2018 variable, with 160 bytes if there is no SYNC segment, or

240 bytes when the SYNC segment is present. All three

are now protected. The error caused STOPOVER ABEND.

-New variable S42SNCONC='CONCURRENT*SYNC I/O*READ+WRITE'

added to TYPE42SR and TYPE42DS datasets.

-Jan 03: another incorrect SRLEN value of 320 protected in

MXG 35.36.

Thanks to Harald Seifert, HUK-COBURG, GERMANY.

Change 35.301 If in your ASUMUOW you defined _LDB2ACC as something

VMXGUOW other than DB2ACCT.DB2ACCT (like DB2.DB2ACCT) and the

Dec 30, 2017 DDNAME did not exist in your LIBNAME or DD statements

ASUMUOW would ABEND with a LIBREF not found. Now the

code looks for this condition, gives you an MXGWARN

message and sets _LDB2ACC to _NULL_ so that ASUMUOW

will run.
Change 35.300 The CPUPHYAD format could fail to be created with error

VMACRMFV messages of overlapped values, when there were multiple

Dec 31, 2017 values of CPUPHYAD (CEC Speed), as when you change the

number of engines for an LPAR; that error caused the

TYPERMFV job to ABEND USER 999.

Thanks to MP Welch, Bank of America, USA.

Change 35.299 Luminex now offers a small Linux appliance called an MDI

MDIADHOC (Mainframe Data Integration) that provides another way to

MDIJCL move MXG jobs off of zOS and onto an ASCII platform. Jobs

MDIJCL1 are still submitted from zOS, so your scheduling system

MDIJCL2 is still in control, but the actual processing of the SMF

MDIJCL3 data occurs on the LINUX platform, and the output PDB's

MDIJCL4 live on storage attached to the MDI. Reports can be sent

MDIPDB back to zOS or routed on your network wherever the MDI

MDIPDB1 can attach. Each job uses 2-3 virtual tape devices, for

MDIPDB2 the SMF input, the SASLOG, and the SASLIST. One site had

MDIPDB3 1TB of SMF, split when SMF was dumped into three outputs,

MDIPDB4 DB2, CICS, for each of 12 LPARS, so 36 concurrent jobs

Dec 31, 2017 processed that data in a bit less than two hours on a

Jan 6, 2018 single MDI.

These sample members provide examples of JCL and code

to run jobs on the MDI. The JCL is very case sensitive

and casing for program names must match the case as

stored in your USERID.SOURCLIB. Also the names and case

of the LOG and LIST datasets in the LUMXPROC must match

the program name.

MDIJCL /MDIPDB runs a basic BUILDPDB adding the 42

and 6156 data to the PDB using

UTILBLDP and BLDSMPDB wirh AUTOALOC

MDIJCL1/MDIPDB1 runs a basic BUILDPDB adding the 42

and 6156 data to the PDB using

UTILBLDP and BLDSMPDB wirh AUTOALOC

and suppressing CICS and DB2 data

MDIJCL2/MDIPDB2 runs VMXGALOC with READONLY=YES

(which suppresss the aging of the

directories) and then runs TYPS110

and CICINTRV.

MDIJCL3/MDIPDB3 runs VMXGALOC with READONLY=YES

(which suppresss the aging of the

directries) and then runs READDB2.

MDIJCL4/MDIPDB4 runs VMXGALOC with READONLY=YES

(which suppresss the aging of the

directries) and after MDIJCL2 and

MDIJCL3 have run will run ASUMUOW.

MDIADHOC JCL for adhoc reporting allows you to

write your program on z/OS and run the

MDI.

Thanks to Chuck Hopf, Independent Consultant, USA.

Thanks to Earl Kline, Luminex, USA

Thanks to Paul Massengill, Luminex, USA

Thanks to Daniel Saunders, Luminex, USA

Thanks to David Feimer, Luminex, USA

====== Changes thru 35.298 are in this MXG 35.12 dated Dec 26, 2017=====
Change 35.298 While all MXG Variables are upper case, mixed case names

VGETFMT are allowed, so you can easily create variable names with

VGETLABL lower case characters, but the listed macros all failed

VGETLEN to find those variable names. The macros are revised to

VGETVAR UPCASE both sides of the compare without changing the

Dec 23, 2017 returned variable name.

Change 35.297 Support for BETA 93 Version 610 (update) and 620 (added).

EXTYBETK -BETA1 blank values for BETADCR corrected and new vars:

EXTYBETL I001PTYPE='PROTOCOL*TYPE'

EXTYBETM I001IPADDR='IP*ADDRESS'

EXTYBETN I001HOST ='HOST*NAME'

FORMATS I001PORT ='HOST*NAME'

VMACBETA I001QUEUE ='HOST*NAME'

VMXGINIT I001FFPARM='HOST*NAME'

Dec 23, 2017 -Support for 620 new subtype 59 creates three datasets and

subtype 22 record is now decoded and creates BETA22VAL:

TYBETK BETA59 SUBTYPE 59 STC START/STOP

TYBETL BETA59RFF SUBTYPE 59 SFF JOB STATS

TYBETM BETA59SFF SUBTYPE 59 SFF JOB STATS

TYBETN BETA22VAL SUBTYPE 22 VALUES

Change 35.296 A WHERE clause in PROC SQL is case sensitive, and if you

VGETLABL create your own variable NAMEs with low case characters,

VGETFMT your variable will NOT be found; only upper case variable

VGETLEN names are found. MXG does not create low case names, and

Dec 20, 2017 cannot detect them in the WHERE clause. No code changed.
Change 35.295 Support for 164-byte DBCTL segment for CICSTRAN dataset.

IMACICDB

Dec 20, 2017

Thanks to Ervin Claxon, CSX Technology, USA.

====== Changes thru 35.294 are in this MXG 35.12 dated Dec 20, 2017=====
Change 35.294 SMF 116 records with MQMSSSID mismatched to QWHSIDMQ

VMAC116 printed MISMATCHED message on the log for every mismatch;

Dec 19, 2017 now, only the first three are printed.

Thanks to Denise Williers, Wipro, USA.

Change 35.293 &MXGCIEXC is a new "exit" for UTILEXCL wherein you can

UTILEXCL correct mispellings in USER CMODHEAD field that cause the

VMXGINIT DUPLICATE CONN report (same offset has two names, usually

Dec 19, 2017 caused by a typo by the CICS SYSPROG who does the DFHCT

assembly of that Monitor Control Table). This change

circumvents the need to reassemble. You would use:

//SYSIN DD *

%LET MXGCIEXC=

%QUOTE(

IF CMODHEAD='PSB ACTV' THEN DO;

PUTLOG _N_= CMODNAME= CMODHEAD=

SMFPSRVR= MCTSSDRL= MCTSSDCN=;

CMODHEAD='PSB ACTI';

END; );

%INCLUDE SOURCLIB(UTILEXCL);

_BLDDICT _BLDEXCL _RPTEXCL

Thanks to Denise Williers, Wipro, USA.
Change 35.292 A utility to Compare numeric variables values in OLD and

UTILCMPR NEW versions of the same SAS dataset, using PROC MEANS to

Dec 30, 2017 compare the value of each statistic of each variable.

%UTILCMPR(IN1=OLD.ZRBASI,IN2=NEW.ZRBASI);

Change 35.291 GCHART AXIS statements were made compatible with WPS

GRAFCEC 03.03.02.00.0222553, and some code was simplified and

GRAFLPAR logic added to STOP if no obs in RMFINTRV and to not

GRAFWRKX plot eligible times if there were none.

GRAFWRKC The WPS graphs are printed in different order, all for a

Dec 15, 2017 particular metric, whereas SAS prints all for an LPAR.

While GRAFLPAR is supported, it is obsolete and GRAFCEC

should be used instead, as it has superior reports.

-As WPS does not support SGPLOT, GRAFWRKC plots had to be

duplicated using the GBARLINE and GCHART procedures and

INCODE= added for data selection.
Change 35.290 Clear of _HSMPLEX macro added at end so that you can

ASUMHSM execute ASUMHSM multiple times in a single job.

Dec 13, 2017
Change 35.289 Support for TYPE42 APARs OA52132, OA52133, and OA61734,

VMAC42 originally coded in Change 35.240, has now been revised.

Dec 12, 2017 A Subtype 5 STOPOVER was caused by new records with the

SRLEN=480 but with actual SR Segment length of 160, or

240 if the new SYNC segment is present, but the actual

length of the SYNC segment is 80 bytes with APAR only

documenting 72. Finally, records with SRLEN=208 and no

SYNC segment are written with only 160 bytes documented.

Change 35.288 Support for new TYPECONF GKEYPM variable and new length

VMACQACS of QAPMDISK of 695 to align those records, although no

Dec 11, 2017 new fields are input in this iteration, awaiting doc.

Thanks to Larry E. Hanus, DST Systems, USA.

Change 35.287 -MXG 35.10 and 35.11 RMF III ZRBASI deaccumulate was WRONG

VMACRMFV DESIGN: should NOT replace ASICPUTA with ASICPUTA_LF, and

Dec 15, 2017 WRONG IMPLEMENTATION: insufficient QA tests, causing the

Dec 21, 2017 value in both variables to be frequently wrong, and if

there were multiple CEC Speeds (CPUPHYAD values) the MSU

value in ZRBASICPUMSU=ZRBASI*CPU MSU*COUNT was wrong.

-This change restores the original ASICPUTA value and the

deaccumulated higher resolution value is in ASICPUTA_LF

so YOU can choose to use the variable of YOUR choice.

-This change also adds variable CPC_CECNAME to ZRBLCP and

ZRBLCPLPAR datasets, and creates a format for CPUPHYAD

lookup (by SYSPLEX SYSTEM) from ZRBCPU to pass CPUPHYAD

into the ZRBASI dataset for MSU calculations.

-Duplicate ASI records for the same task in an interval do

exist, as when a task changes it's JOB name, and they are

visible in RMF III reports, but the deaccumulate can be

a missing value as IBM has not provided a way to identify

which was the first observation and which was the second.

In addition, records with seconds of CPUTCBTA value and

microseconds for CPUTCBTA_LF have been observed, so the

value of using CPUTCBTA_LF needs to be examined in your

data. These issues are open with RMF development, and

this text will be updated when more is known.

-Dec 21: Invalid INPUT for PHYCPUAD message had no impact

but was corrected; was printed when no ZRBCPU matched.
Thanks to MP Welch, Bank of America, USA.
Change 35.286 MXG variable IOTMNOCA, uncaptured IO Connect Time in 30s,

BUILD005 was incorrectly calculated in BUILDPDB and SMFINTRV using

BUIL3005 SMF30AIC-IOTMTOTL instead of -IOTMDASD causing negative

SMFINTRV values. But SMF30AIC is the connect time for the ASID

Dec 6, 2017 and Dependent Enclaves, but does NOT include FICON chans

which could also cause negative values.

Thanks to Randy Hewitt, DXC Technology, USA.
Change 35.285 Support for Container Pricing in RMF 72 records creates

EXTY72TR new TYPE72TR dataset for Tenant Resource Group that are

IMAC7072 added by APAR OA52694. TYPE72TR has the same variables

VMAC7072 that are in TYPE72GO with new variables for TRG

VMAC79 R723GGTI='TENANT*IDENTIFIER'

VMXGINIT R723GGTN='TENANT*NAME'

Dec 7, 2017 R723GGKY='TENANT*SOLUTION*ID'

R723GGTF='TENANT*RESOURCE*GROUP?'

and these variables added to both TYPE72GO and TYPE72TR:

R723CPA_ACTUAL ='PHYSICAL*CPU*ADJUSTMENT*FACTOR'

R723CPA_SCALING='SCALING*FACTOR*FOR*R723_ACTUAL'

-Flag variables added to TYPE792 and TYPE795 dataset

R792FLG32='R792RGRP*IS A*TRG?'

R795FLG6='R795RGRP*IS A*TRG?'

Change 35.284 -MXG 35.11 inserted statements to create ZRBCPUxxxMSU vars

VMACRMFV incorrectly inside MACRO _EZRBCPU and _EZRBLCP definition

Dec 6, 2017 causing UNINIT variable if you had tailored EZRBCPU.

Dec 8, 2017 -Missing values for ZRBCPUZIPMSU were corrected. The new

Dec 9, 2017 ZRBLCPLPAR dataset requires the CPUG3 CPCDB, and SSHG3

tables, or ASMRMFV Table IDs of CPU and CPC, since SSH is

always selected.

Thanks to MP Welch, Bank of America, USA.

Change 35.283 Support for APAR OA52828 which allows customization of

VMAC0203 the size of the SMF Temporary Buffer used to hold SMF

VMAC7 data during IPL processing.

VMAC23 TYPE0 dataset new variable

Dec 5, 2017 SMF0TBUF='SMFTBUFF*PARAMETER*SPECIFIED*MEGABYTES'

TYPE7 dataset new variable

SMF7TBLS='BYTES LOST*DURING SMF*INITIALIZATION';

TYPE23 dataset new variable

SMF23MBU='MAX BYTES*STORED IN*SMFTBUFF'
Change 35.282 A new set of variables added to TYPE70 and RMFINTRV to

VMAC7072 capture the number and usage of IIP, IFL, and ICF CPS

VMXGRMFI for the platform (CEC).

Dec 5,2017 PLATZIPCPUS - The number of IIPs on the CEC

PLATZIPBUSY - The total % busy of all the IIPs

PLATIFLCPUS - The number of IFLs on the CEC

PLATIFLBUSY - The total % busy of all the IFLs

PLATICFCPUS - The number of ICFs on the CEC

PLATICFBUSY - The total % busy of all the ICFs
Change 35.281 TACI802 dataset variable FINTIME was not converted from

VMACMVIP GMT to LOCAL Time zone.

Dec 4, 2017

Thanks to Paul Volpi, UHC, USA.

Change 35.280 The Exit member EXDB2STS and _EDB2STS macro are now used

VMACDB2 to give control of the output of PDB.DB2STATS; previously

Dec 6, 2017 defined but not used.

Thanks to Scott Barry, SBBWorks Inc., USA.

====== Changes thru 35.279 are in this MXG 35.11 dated Dec 1, 2017=====
Change 35.279 Support for Dec 2017 z14 CPU MF formula update.

ASUM113 The EXTND158 counter was moved from L3P to L4LP.

VMAC113 -John's updated formulas are available at

VMACVMXA

Nov 30, 2017

http://www.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/TC000066

Change 35.278 Correction for TMS Stacked Tape Files; these variables

TYPETMS5 were not retained from the first DSNB record into the

VMACTMS5 "CHANGED" records so they were incorrect:

Nov 30, 2017 RFILSEQ RLRECL RBLKSIZE RRECFM RSTPNAME RFILPERC;

Thanks to Scott Barry, SBBWorks Inc., USA.
Change 35.277 Support for IFCID=225 new fields in DB2STAT4 dataset:

VMACDB2 QWA225PRISTG_PAGE /**DBM1*PRVT ELIG*PGSTEAL*/

Nov 28, 2017 QWB225PRISTG_PAGE /**DIST*PRVT ELIG*PGSTEAL*/

Dec 6, 2017 IBM DSECT notes these fields contain: