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

//SYSIN DD *

%LET MACMVCH=

%QUOTE(

IF MVCIDSN=' ' THEN DO;

MVCIDSN=STRIP(SCAN(MVCIJFCB,1,' ');

MVCIGDG=SCAN(MVCIDSN,-1);

RETAIN MVCIDSN MVCIGDG;

END;

CALL SYMPUT('MVCIDSN',MVCIDSN);

);

%INCLUDE SOURCLIB(TYPEMVCI);

%PUT INPUT DSNAME WAS: &MVCIDSN;

would create variables MVCIDSN,MVCIGDG, and would create

a macro variable &MVCIDSN with the DSNAME that was read.

Thanks to Henk van der Veur, Fortis, THE NETHERLANDS.

Change 26.313 Revisions to support for TMON for DB2 V4 and V4.1.

VMACTMDB -Dataset TMDBDB2 had misalignments corrected.

Jan 23, 2009 -Support for BF variables in dataset TMDBBF.

-Support for BG variables in dataset TMDBBG.

-Support for BH variables in dataset TMDBBH.

-Support for BI variables in dataset TMDBBI.

Thanks to Ernie Amador, University of California Davis Health, USA.
Change 26.312 Support for APAR OA27161 which adds interval variables

VMAC23 to the SMF 23 record. Change 26.116 supported OA22414,

Jan 22, 2009 which added cumulative variables for the same metrics.
Change 26.311 -In DB2 V9, IFCID=225 creates the DB2STAT4 dataset and

READDB2 DB2 V8 creates the T102S255 dataset, but DB2STAT4 had

VMACDB2 all missing values for the QW0225xx variables in the

Jan 22, 2009 DB2STAT4 dataset because QWS02R1O was spelled with a

Jan 28, 2009 zero instead of an oh in VMACDB2.

Feb 2, 2009 -READDB2 could cause error messages

THE TEXT EXPRESSION LENGTH (65545)

EXCEEDS MAXIMUM LENGTH (65534).

-READDB2 did not create both T102S225 and DB2STAT4 when

IFCIDS=225 was specified, in spite of the claim made in

Change 26.233. Now, both datasets are created.

-Because T102S225 is created from SMF ID=102 but the

DB2STAT4 is created from SMF ID=100 subtype 4, there

are still some limitations in READDB2 using WANTONLY.

-New-in-DB2-V9 variables added and kept in DB2STAT4:

QW0225SF='FIXED*VIRTUAL*64BIT*SHARED'

QW0225SG='GETMAINED*VIRTUAL*64BIT*SHARED'

QW0225SV='VARIABLE*VIRTUAL*64BIT*SHARED'

QW0225SU='STACK*STORAGE*IN USE'

QW0225L2='ALLOC STORAGE*CACHED*THREADCPYS*ABOVE POOLS'

QW0225H2='HWMALLOCSTORE*CACHED*THREADCPYS*ABOVE POOLS'

QW0225F1='SERVICEABILITY*QW0225F1'

QW0225F2='SERVICEABILITY*QW0225F2'

QW0225S2='STATEMENT*CACHE*BLOCK*STORAGE*GT 2GB'

-This example shows how you can use MXG macros instead

of using READDB2 if you only wanted the DB2 Statistics

datasets (DB2STATS/DB2STATB/DB2STATR/DB2STAT4/DB2GBPAT

DB2GBPST) and T102S225 (i.e., you have both V8 and V9

IFCID=225 records in your SMF); the example reads only

the SMF 100 and 102 records and does not create any of

the DB2ACCTx datasets.

%LET MACFILE=

%QUOTE( IF ID=100 OR ID=102; );

%LET MACKEEP=

_N102

MACRO _W102225 PDB.T102S225 %

MACRO _EDB2ACC %

MACRO _EDB2ACB %

MACRO _EDB2ACP %

MACRO _EDB2ACG %

MACRO _WDB2ACC _NULL_ %

MACRO _WDB2ACB _NULL_ %

MACRO _WDB2ACP _NULL_ %

MACRO _WDB2ACG _NULL_ %

;

%LET PDB2ST0=WORK;

%LET PDB2ST1=WORK;

%LET WDB2PAT=PDB;

%LET WDB2PST=PDB;

%LET WDB2STR=PDB;

%LET WDB2ST2=PDB;

%LET WDB2ST4=PDB;

%INCLUDE SOURCLIB(VMACDB2,VMAC102,IMACKEEP);

DATA _VARDB2 _V102225;

_SMF

_CDEDB2

_HDR102 _C102225 _END102;

_SDB2STB

_SDB2STS;

RUN;

Thanks to Rachel Holt, Fidelity Systems, USA.

Thanks to Dan Case, Mayo Clinic, USA.

Change 26.310 The MXGWORK= argument of VMXGINIT hasn't worked for many

VMXGDEL years (Change 18.148, Jun 2000, Change 17.060, 1999) as

VMXGINIT a way to change the //WORK destination, and it was the

Jan 21, 2009 cause of errors reported in Change 26.252, so it has

been removed as an argument and is now a GLOBALed macro

variable name in VMXGINIT, and the now-redundant GLOBAL

statement in VMXGDEL is removed.

Thanks to Chris Weston, SAS ITRM Development, USA.

Change 26.309 Variable TMNTJOBC='JOB*CANCELLED*BY*OPERATOR' was wrong

VMACTMNT and always was the same as TMNTJOBI, because TMNTJOBC

Jan 21, 2009 was set 'Y' from the x'02' bit instead of the x'01' bit.

Thanks to Michael Creech, Lender Processing Services, USA.

====== Changes thru 26.308 were in MXG 26.12 dated Jan 20, 2009=========
Change 26.308 The TYPE70 dataset is enhanced with these counts of the

VMAC7072 specialty engines:

Jan 17, 2009 PARTNICF='PARTITION*NUMBER OF*ICF*ENGINES'

PARTNIFA='PARTITION*NUMBER OF*ZAAP*ENGINES'

PARTNIFL='PARTITION*NUMBER OF*IFL*ENGINES'

PARTNZIP='PARTITION*NUMBER OF*ZIIP*ENGINES'

Thanks to Jim Horne, Lowe's Companies, USA.
Change 26.307 The CLIST to process RMF III files had an incorrect GOTO

CLRMFV in line 943 that should be GOTO DONE. Fortunately, this

Jan 16, 2009 path thru the CLIST is seldom-to-never used.

Thanks to Jerry Urbaniak, Acxiom, USA.

Change 26.306 -If the new operand COPYONLY=YES was specified, READDB2

READDB2 tried to read the header using VMACDB2H, but that read

Jan 15, 2009 required prerequisites. Code now uses _CDEDB2 which

then calls VMACDB2H with all datasets to _NULL_ via the

_NDB2 to read only the headers if COPYONLY=YES.

-READDB2 was enhanced to allow selection by QWHCEUTX,

the End User Transaction Name, which might be useful

in the classification of DDF work in WLM, but only

if it is filled in by the calling application with

something other than a default value. It must also be

the first piece of the unit of work that is seen by WLM.

Thanks to Becky Clark, Bank of America, USA.

Change 26.305 CICSTRAN variables RTYPE and RRTYPE contain $EBCDIC text

VMAC110 but some of the INPUT statements in VMAC110, and in code

UTILEXCL built by UTILEXCL, incorrectly had $CHAR inputs. As long

Jan 15, 2009 as MXG is executed on z/OS, there was no difference, but

for ASCII execution, the INPUTs must be $EBCDIC.

Thanks to Glenn Bowman, Wakefern, USA.

Change 26.304 Internal changes for DOCVER enhancements and CONFLICTs.

FORMATS New $MGUTILD format maps each MXG-built DATASET name in

UTILVREF member DOCVER to both the INFILE that was read, and the

Jan 14, 2009 MXG member name that created that dataset, and a flag

for "second-level" datasets that are not created from an

infile but are created from SAS datasets. The format is

now used to add that information in the DOCVER member.

Also, the format is used in the new QA CONFLICTs report

to cluster datasets by INFILE to eliminate conflicts in

variable's type, length, format, or maybe even labels.

-So, if FORMAT $MGUTIL has been updated, but the QA Job

has NOT been rerun, the new datasets in DOCVER will NOT

have their INFILE identified; that line will be blank.
Thanks to Freddie Arie, Merrill Consultants, USA.
Change 26.303 Capture Ratios for zAAPs and zIIPs are now calculated in

VMXGRMFI PDB.RMFINTRV and TREND.TRNDRMFI datasets, and the total

Jan 13, 2009 Service Class zAAP and zIIP CPU times are creates:

CPUIFATM - IFA (zAAP) CPU TIME in Service Classes

CPUZIPTM - ZIP (zIIP) CPU TIME in Service Classes

IFAACTTM - IFA (zAAP) CPU TIME in TYPE70 (Hardware)

ZIPACTTM - ZIP (zIIP) CPU TIME in TYPE79 (Hardware)

CAPIFART - CAPTURE RATIO FOR ZAAP ENGINES

100*CPUIFATM/IFAACTTM

CAPZIPRT - CAPTURE RATIO FOR ZIIP ENGINES

100*CPUZIPTM/ZIPACTTM

Thanks to Brian Harvey, HCL America, USA.

Change 26.302 INPUT STATEMENT EXCEEDED for unexpected AAA record with

VMACNMON AAA,LPARNumberName,none

Jan 13, 2009 MXG expected LPARNumberName to be followed by the LPAR

number and the LPAR name. Why the LPARNumberName tag

exists when its value is "none" is unknown, but now, the

LPARNAME='none' will be output.

Thanks to Steven Olmstead, Northwestern Mutual, USA.
Change 26.301 Debugging PUT statements were always printed, because

UDB2GTF the IF DEBUG GE 2 THEN statement did not precede PUT.

Jan 13, 2009

Thanks to Steve Wood, DST Systems Inc, USA.

Change 26.300 Conflict Resolution of variable's TYPE/LENGTH/FORMAT.

VMAC94 -In MXG 26.11, I made changes to eliminate exposures to

VMAC89 possible conflicts (none had actually occurred), but

VMACTPMX did not document all of the changes in Change 26.289.

DOC One impacting change is now reversed: variable PRODREL

Jan 18, 2009 in TYPE892 has been restored (it was renamed PRODRL89 in

MXG 26.11, but that impacted the LCS product); it length

is now $6 to match other PRODREL variables, eliminating

any exposure.

-In MXG 26.11, TYPETPMX variable ACCT was renamed ACCTJOB

to avoid conflict with the older TYPE26J2 ACCT variable,

but incorrectly, causing an VARNAME ACCTJOB NOT FOUND.

Since ACCTJOB is then stored into the actual ACCOUNTn,

it should probably not have been kept, and is unlikely

that the change from ACCT to ACCTJOB has any impact.

But the new ACCTJOB variable is length $16 and all 16

converted into $EBCDIC, so individual ACCOUNTn fields of

16 are now supported; actual length is set in IMACACCT.

-In TYPE94, an actual conflict was reported by Ken and is

now fixed: the format of variables STARTIME and DURATM

was DATETIME18.0 and TIME8., because the SMF 94 records

only have seconds resolution; however, adding type 94 to

your BUILDPDB caused all of the RMF datasets to have the

shorter format for their STARTIME and DURATM variables,

as SAS uses the last instance of a FORMAT statement. By

changing the FORMAT in VMAC94 to match the "standard"

DATETIME21.2 and TIME12.2 formats the full RMF precision

is printed.

However, this is truly ONLY cosmetic; the FORMAT is

only used when values are printed, and they do NOT

change the stored data values; you can ALWAYS use a

FORMAT statement in your PROC PRINT, etc., to show

as many or as few decimals as you want displayed.

-The below variables were renamed to eliminate conflicts.

Many of these datasets are no longer even creatable, and

none are created from common IBM SMF records - most are

created from optional user SMF records and none of these

renamed variables are likely to be used in your reports.

Dataset Variable Name Dataset Variable Name
AIM098_R was FILENAME SAMONASR was APPLNAME

now FILENAAI now APPLNASA

CCCDAT was ACCOUNT SAMONAUR was APPLNAME

now ACCTCCC now APPLNASA

CMHMEVNT was EXPDT SAMONTSR was APPLNAME

now EXPDTDAT now APPLNASA

FILTEKID was GROUP SAMONUSR was APPLNAME

now GROUPSM1 now APPLNASA

ICEBRGDE was RELEASID SUPERIND was TERMINAL

now RELEASIC now TERMSUIN

ICEBRGUT was VDEVTYPE was USERNAME

now VDEVTYIC now USNASUIN

ILKVCONN was USERDATA SV08THRE was ABEND

now USDAILKV now ABENDTSK

ILKVDISC was USERDATA was CLASS

now USDAILKV now CLASSVIE

LDMSBASE was FUNCTION now GROUPTH3

now FUNCLDMS was GROUP

LDMSDIST was FUNCTION was STATUS

now FUNCLDMS now STATSVIE

LDMSPC was SERVER SV25TSUM was RESPTIME

now SERVLDMS now RESPSVIE

MEMOACCT was ACCTCODE SV27TRAN was NETNAME

now ACCTCOME now NTNMSVIE

was SESSION was OFCTYNME

now MEMOSESS now OFCTYTCP

was USERNAME was RTYPE

now USNAMEMO now RTYPSVIE

MVTPN was MTU was TRANPRI

now MAXMTU now TRANPRIS

MVTPS was LOCATION was TERMINFO

now LOCNMVTP now TERMINFS

was NAME T112MQCO was MQTRAN

now NAMEADMN now MQTRANID

NAFENTVA was TERMINAL T112MQCT was MQTRAN

now NAFTERM now MQTRANID

NAFGPSTO was VLU TYPE1031 was SUBSYS

now VLUNAF now SUBSY103

NAFGPSTR was VLU TYPE1032 was SUBSYS

now VLUNAF now SUBSY103

OMCIADA was AFNAME TYPE6367 was ACTION

now AFNAMECI now ACTIONRQ

OMCIBOTL was ESTTIME TYPEACC was MESSAGE

now ESTDTIME now MESSGACC

OMCIFIXD was ESTTIME TYPEDLMN was NEWNAME

now ESTDTIME now NWNMDLMN

OMCIRTA was ESTTIME was USERNAME

now ESTDTIME now USNADLMN

OMCITITL was ESTTIME TYPESTRS was VERSION

now ESTDTIME now VERSSTRS

OMCITRAL was ESTTIME TYPETPMX was ACCT

now ESTDTIME now ACCTJOB

OMCITRAN was ESTTIME TYPEX37 was ACCTNO

now ESTDTIME now ACCTNO37

OMCIVDCT was ESTTIME was MESSAGE

now ESTDTIME now MESSGX37

OMCIVENQ was ESTTIME was PRODREL

now ESTDTIME now PRODRL37

OMCIVFCT was ESTTIME XCOMDATA was USERID

now ESTDTIME now USERXCOM

OMCIVIO was ESTTIME XPTR10 was ACCOUNT

now ESTDTIME now ACCTXPTR

OMCIVJCT was ESTTIME XPTR21 was ACCOUNT

now ESTDTIME now ACCTXPTR

OMCIVPCP was ESTTIME was LOCATION

now ESTDTIME now LOCNXPTR

OMCIVVSA was ESTTIME XPTR45 was ACTION

now ESTDTIME now ACTNXPTR

PRORECOV was PRODREL ZARAERRV was VOLPOOL

now PRODRLPR now VOLPOOLZ

ROSCOAUD was USERID ZARAVOL was VOLPOOL

now USERROSC now VOLPOOLZ

ROSCOE was USERID

now USERROSC

was TERM

now TERMIOS

ROSCORPS was USERID

now USERROSC

ROSCOSTA was USERID

now USERROSC

ROSCOVPE was USERID

now USERROSC

Thanks to Al Sherkow, I/S Management Strategies, Ltd.

Thanks to Scott Wigg, U.S. Bank, USA.

Thanks to Kenneth D. Jones, Bell Aliant, CANADA.
Change 26.299 MXG 26.10-26.11. PERFINDX=. for all service classes with

VMAC7072 R723TYPE=2:ADDRESS SPACE WITH NO TIME GOAL. A "cosmetic"

Jan 12, 2009 Change 26.269, eliminated "missing value" value messages

but accidentally setting PERFINDX to a missing value for

those service classes.

Thanks to Dan Melton, Lowe's Companies, USA.

Change 26.298 DB2 ID=102 IFCID=22 INPUT STATEMENT EXCEEDED RECORD

VMAC102 error when length in the header field was zero; the

Jan 11, 2009 actual length is 2 bytes longer than stored in the

start of each segment. MXG logic revised to support.

Thanks to Tom Buie, Southern California Electric, USA.
Change 26.297 -Case in zPCR and ExternalSource tags were inconsistent

ANALZPCR and caused configuration errors.

Jan 13, 2009 -SELECT=CECTIME created multiple HOST tags.

-Printed ONLPCPUS in reports was correct, but the LCPs

tag value could be one higher than true number of CPs

in the LPAR, if ONLPCPUS was an integer. MXG algorithm

ONLPCPUS=CEIL(ONLPCPUS) now ONLPCPUS=CEIL(ONLPCPUS-.01)

to protect for both 3.0 and 2.999 values, and relocated

so the printed and tag values are the same.

-An "External Study File" can ONLY be LOADed from zPCR's

FUNCTION SELECTION window, using its FILE pulldown LOAD

option to browse/select the MXG-built xxxx.zpcr file.

Thanks to David Bixler, FISERV, USA.
====== Changes thru 26.296 were in MXG 26.11 dated Jan 5, 2009=========
Change 26.296 The BLSR option for IMS VSAM files is not supported; the

ANALBLSR ANALBLSR analysis created pages of reports for IMS VSAM

Jan 5, 2009 files, with PROGRAM='DFSRRC00', so this change deletes

all files that were accessed with that program name.

Thanks to Stephen Hughes, Excellus, USA.
Change 26.295 The RMFWKLRV interval workload dataset build by RMFINTR

VMXGRMFI creates two new variables WKLDIOCN and WKLDIORT, the

Jan 2, 2009 I/O count and I/O rate from R723CIRC.

Thanks to Don Goulden, SAS ITRM Development, USA.

Change 26.294 Two new arguments, CICSTRAN & DB2ACCT can be specified

UTILBLDP in the SUPPRESS= argument if you do not want the detail

Jan 2, 2009 CICSTRAN or any of the DB2ACCTx datasets to be created.

With SUPPRESS=CICSTRAN, SMF 110 subtype 1 records are

skipped, with SUPPRESS=DB2ACCT, SMF 101 records are.
Change 26.293 %VMXGPRAL(DDNAME=PDB,NOBS=20) prints ALL datasets in a

VMXGPRAL SAS data library, but now, it uses new VMXGPRNT to print

VMXGPRNT both the variable's LABEL and the variable's NAME in the

Jan 2, 2009 heading. The new VMXGPRNT macro can also be used alone

to print a single SAS dataset, or it can be used to

create a Comma-Separated-Variable file from a single SAS

dataset. I find VMXGPRAL very useful to print a few obs

with all of the variables to validate a new MXG-built

dataset. The logic for VMXGPRNT was originally posted

to MXG-L by John, and Mike recently reminded me of it.

The old UPRINDOC member did print the label and name,

and is used to create the ADOCxxxx prints/means, but

the new VMXGPRAL and VMXGPRNT is much nicer syntax.

Thanks to Mike O'Brien, Bank of America, USA.

Thanks to John Parkes, Experian, USA.
Change 26.292 MONTHBL3 was overlooked when Change 24.064 added simpler

MONTHBL3 override for NOT SORTED errors in creating MONTH PDB.

Jan 1, 2009 It is now updated to match MONTHBLD logic.

Thanks to Douglas G. Wells, First National Bank of Omaha, USA.

Change 26.291 WPS does not support the INFILE CCHHR nor CCHHR= options

VMACEREP which is required to read LOGREQ with TYPEEREP, and was

VMXGVTOC used in VMXGVTOC (now archaic, replaced by DCOLLECT), and

UTILSPAC also used in the UTILSPACE reporting utility. This

Jan 1, 2009 change removes the CCHHR options when MXG is executed

under WPS, so that the QAWPS QA test stream can run, to

test the compiler using only DUMMY INFILES. This change

will be revised if/when CCHHR support is in WPS.

See Change 27.239.
Change 26.290 The View Direct, Mobius, Infopac-RSD, etc., product's

VMACIPAC user SMF record, subtype 3 was INCOMPATIBLY restructured

Dec 30, 2008 with R3CALLER field relocated, the LUNAME field removed,

and seven extra bytes inserted. MXG was recoded to read

the new format with a heuristic test of the old location

of R3CALLER, pending response from [e-mail], who

now owns this product.

Jan 5: A PTF has been provided, but not yet installed,

but even if it does correct the invalid SMF record, I

will leave the circumvention in place, since it will

protect sites without the correcting ptf.

Thanks to Jim Horne, Lowe's Companies, USA.

Change 26.289 -Cleanup of inconsistent LENGTHs and CHAR/NUM conflicts

ANALCISH for variables created from the same INFILE. CHAR/NUM

VMAC112 conflicts were resolved by renaming the variable from

VMACFTEK the more obscure product. Lengths were mostly increased

VMACQACS with &MXGLEN replacing 4s, but some 8s are now &MXGBYLN.

DIFFROSC These LENGTH changes may cause WARNINGs with SAS V9.2,

VMACSVIE when creating WEEK or MONTH PDBs with PDBs built by MXG

VMXG70PR versions before/after this change, but by installing now

VMACSVIE if still under SAS V9.1.3, then your migration to V9.2

VMAC7072 could be warning-free.

No CHAR/NUM conflicts had been reported, so changes were

Jan 5, 2009 made to prevent future errors.

-Data Set Labels were added where they were blank.

-Variable ZDATE was added where it was not being created.

-Member DOCVER26 lists all of the changed variable names:

Variables TERMINFO TRANPRI were Numeric in SV27TRAN data

set built by TYPESVIE from SMF, which conflicted with

IBM SMF 110 records, so TYPESVIE was changed to create

them as character to avoid conflict if 110s and SVIE

were ever (unlikely) combined in a single DATA step.

ANALCISH, typo, CICSCONSR changed to CICCONSR.

VMACFTEX, variable GROUP renamed to GROUPSM1.

VMAC112, variable MQTRAN renamed to MQTRANID.

DIFFROSC, variable TERM renamed to TERMIOS.

VMACQAPM, variable XIDTYP changed to CHAR.

VMXG70PR, variable IFAUPTM LPARNUM SMF70CSF PCTIFABY

PCTZIPBY changed from length 8.

VMAC7072, variables NRPRCS, NRPHYCPS changed from len 8.

Change 26.288 TYPE78CU variables for Aliases could be wrong, as they

VMAC78 were retained but not initialized, so devices without an

Dec 29, 2008 alias could incorrectly have values in these variables:

R783HCNT R783HIX R783HLCU R783HCU R783HNAI R783HTIO

R783HAIU R783HCAD R783HIOQ

Thanks to Peter Schubert, CITEC, AUSTRALIA.

Change 26.287 Support for RMF III CPUG3 z/OS 1.9 (INCOMPATIBLE - zero

VMACRMFV observations in ZRBLCP without this change).

Dec 28, 2008 New variable in ZRBLCP: LCPUCURW - Current SHARE weight.

LCPUCURW='CURRENT*SHARE*WEIGHT'

New variables in ZRBCPU:

CPCGRPNM='CAPACITY*GROUP*NAME'

CPCGRPLM='CAPACITY*GROUP*MSU*LIMIT'

CPGRPJOI='DATETIME WHEN*LPAR JOINED*GROUP'

Thanks to Stephen Hoar, Lloyds TSB, ENGLAND.
Change 26.286 Very strange error messages in 26.08 and 26.10, where

VMXGRMFI the "x" is a non-printable character:

Dec 23, 2008 ERROR: LIBNAMES ARE RESTRICTED TO EIGHT CHARACTERS.

x HAS BEEN TRUNCATED.

Removal of an unneeded %VGETENG call in VMXGRMFI has

eliminated the strange errors, but still researching.

Thanks to Dan Case, Mayo Clinic, USA.
Change 26.285 Enhanced to do the dynamic allocation of DAYS based on

VMXGALOC the DAYSKEEP= value. On ASCII this creates a pseudo-GDG

Dec 23, 2008 architecture were days a DddMMMyy, weeks are Wddmmyy and

Jan 3, 2009 months are Mddmmmyy, and a variable number of each can

be kept, with old ones being deleted and new ones being

created over time.

So, if today is 22DEC08, the directory names would be

PDB - c:\MXG\D21DEC08

MON - c:\MXG\D15DEC08

TUE - c:\MXG\D16DEC08 ....

WEEK - c:\MXG\W15DEC08

WEEK1 - c:\MXG\W15DEC08

WEEK2 - c:\MXG\W08DEC08

WEEK3 - c:\MXG\W01DEC08 ....

MONTH - c:\M01DEC08
In addition, TREND databases can be allocated based on

doing TRENDing on a daily or weekly basis.

TRENDING=DAILY creates

TRENDIN - c:\MXG\T20DEC08

TREND - c:\MXG\T21DEC08
TRENDING=WEEKLY creates

TRENDIN - c:\MXG\T08DEC08

TREND - c:\MXG\T15DEC08
TRENDING=WTD creates

TRENDIN - c:\MXG\T15DEC08

TREND - c:\MXG\T22DEC08
The SPIN and TREND directory management follows the

pattern specified in DAYSKEEP WEEKKEEP. Older

directories will be erased/deleted based on the

numbers in those two arguments.

For those times when you may need to go back and rerun a

day's processing, FORCEDAY=22DEC08 will set the correct

directories for that day's processing to be allocated.
The basedir= argument defaults to c:\mxg\ but can be any

disk device and directory available to your system.

Change 26.284 The BMC Optional CMRDB2 segment was increased from 172

IMACICMD to 256 bytes in CICS/TS 3.2. The IMACICMD code is now

Dec 22, 2008 updates to support both lengths, based on SMFPSRVR.

Thanks to Jane Dickenson, Produban, ENGLAND.

Change 26.283 -zPCR failed with PDB=SMF, or SCP not 'z/OS V1R10', the

ANALZPCR SELECT=CECTIME option is now implemented, and new notes

Dec 21, 2008 have been added: