COPYRIGHT (C) 1984-2023 MERRILL CONSULTANTS DALLAS TEXAS USA
MXG CHANGES 37.37
=========================MEMBER=CHANGE37================================
/* COPYRIGHT (C) 1984-2020 MERRILL CONSULTANTS DALLAS TEXAS USA */
MXG VERSION 37.37 is dated Jan 6, 2020, thru Change 37.272.
MXG VERSION 37.09 was dated Dec 20, 2019, thru Change 37.268.
EarlyA MXG VERSION 37.09 was dated Dec 20, 2019, thru Change 37.267.
MXG VERSION 37.08 was dated Nov 26, 2019, thru Change 37.256.
MXG VERSION 37.07 was dated Oct 22, 2019, thru Change 37.239.
Third MXG VERSION 37.07 was dated Oct 14, 2019, thru Change 37.236.
Second MXG VERSION 37.07 was dated Oct 12, 2019, thru Change 37.235.
First MXG VERSION 37.07 was dated Oct 9, 2019, thru Change 37.234.
MXG VERSION 37.06 was dated Aug 30, 2019, thru Change 37.190.
First MXG VERSION 37.06 was dated Aug 22, 2019, thru Change 37.184.
MXG VERSION 37.05 was dated Jul 8, 2019, thru Change 37.154.
Second MXG VERSION 37.05 was dated Jul 6, 2019, thru Change 37.153.
First MXG VERSION 37.05 was dated Jul 5, 2019, thru Change 37.152.
MXG VERSION 37.04 was dated Jun 5, 2019, thru Change 37.124.
MXG VERSION 37.03 was dated Apr 19, 2019, thru Change 37.091.
MXG VERSION 37.02 was dated Mar 11, 2019, thru Change 37.057.
Updated MXG VERSION 37.01 was dated Feb 3, 2019, thru Change 37.031.
First MXG VERSION 37.01 was dated Feb 1, 2019, thru Change 37.029.
Annual MXG Version 36.36 was dated Jan 4, 2019, thru Change 36.255.
The Last MXG Newsletter SIXTY-NINE was dated Jan 3, 2018.
New TECHNOTES for NEWSLTRS are now in CHANGESS.
Instructions for ftp download can be requested by using this form:
http://www.mxg.com/Software_Download_Request
Your download instructions will be sent via return email.
Contents of member CHANGES:
I. Current MXG Software Version 37.37 is available upon request.
II. SAS Version requirement information.
III. WPS Version requirement information.
IV. MXG Version Required for Hardware, Operating System Release, etc.
V. Incompatibilities and Installation of MXG 37.37.
VI. Online Documentation of MXG Software.
VII. Changes Log
Member NEWSLTRS contains Technical Notes, especially APARs of interest
and is updated with new notes frequently. All Newsletters are online
at http://www.mxg.com in the "Newsletters" frame, although there are
no new NEWSLTRS updates; they are now found in CHANGESS as TECHNOTEs.
Member CHANGES contains the changes made in the current MXG version.
Member CHANGESS contains all changes that have ever been made to MXG.
All MXG changes are also online at http://www.mxg.com, in "Changes".
========================================================================
I. MXG VERSION 37.37 DATED Jan 6, 2020, THRU CHANGE 37.272.
==MAJOR CHANGES ADDED IN MXG 37.37, DATED Jan 6, 2020 THRU 37.272.
BUILDIMS 37.271 %BUILDIMS creates all possible IMS data plus report.
CONFIMXG 37.267 DO NOT USE ENCODING=EBCDIC1047, use OPEN_ED-1047.
Data values are corrupted with EBCDIC1047.
ADOCABND 37.270 A TECHNOTE documenting recovery from MXG ABENDS,
TYPE71 37.269 TYPE71 variables SMF72PIS/POS now deaccumulated.,
==MAJOR CHANGES ADDED IN MXG 37.09, DATED Dec 20, 2019 THRU 37.268.
SAS OPTION ISSUE FOR Z/OS ONLY, FOR NONLSCOMPATMODE ISSUES:
which is the z/OS default. EBCDIC1047 corrupts output SAS dataset.
ABENDS CORRECTED
TYPE42 37.261 SMF 42 Subtype 1 TYPE42DS INPUT STATEMENT EXCEEDED.
ASUMMSUS 37.265 ERROR PARM OPTNE NOT DEFINED, typo.
NEW SUPPORT
TYPE99 37.264 SMF 99 Subtype 2 TYPE99_2 new variables added.
TYPEIMS 37.259 IMS SYNCPOINT Log Records IMS5937/5938 keeps all vars
TYPEIDMS 37.260 IDMS dataset IDMSTAS USER Fields were misaligned.
==MAJOR CHANGES ADDED IN MXG 37.08, DATED Nov 26, 2019 THRU 37.256.
TYPE42 37.250 SMF 42 ABEND due to SRLEN=208, now all protected.
VMACSMF 37.249 Using FTP ACCESS from z/OS fails with JFCB issue.
ERRORs CORRECTED
TYPERMFV 37.246 Variables PCIFTET/PCIFTQT in ZRBPCI incorrect.
VMXGSUM 37.245 KEEPALL=NO with INCODE= could cause UNINIT message.
TYPE82 37.241 Several MG082xx formats had wrong hex value.
TYPE102 37.240 Variable QWP4STPGS=STASTPGSAMP/N/Y/S is now INPUT.
NEW SUPPORT
TYPE79 37.242 Support for BMC SMF 70 Subtype 255 creates TYPE70FF.
TYPE70PR 37.243 Vars SMF70LACM/LACA/LACB are now kept in TYPE70PR.
ASMRMFV 37.255 RMF III FDF support for PCIE and SCMG3 tables.
ENHANCEMENT
TECHNOTE 37.247 Example RACF analysis, why USERID was revoked.
TYPE119 37.244 Subtype 3 SMF 119 not written, USERID no SMF access.
DIFFROSC 37.256 New ROSCOE dataset ROSCOMON with all MONITORS.
==MAJOR CHANGES ADDED IN MXG 37.07, DATED Oct 22, 2019 THRU 37.239.
ERROR CORRECTED
TYPE113 37.239 Support for new SMF 113 CRYPTO17-20 for z/15,
caused ABEND: ARRAY SUBSCRIPT OUT OF RANGE,
only if HIS CRYPTO counters are enabled, and.
only on z/15.
==MAJOR CHANGES ADDED IN MXG 37.07, DATED Oct 14, 2019 THRU 37.236.
ERROR CORRECTED
TYPE113 37.236 Support for new SMF 113 EXTND255-287 for z/15.
Caused ABEND: ARRAY SUBSCRIPT OUT OF RANGE,
only on z/15.
==MAJOR CHANGES ADDED IN MXG 37.07, DATED Oct 12, 2019 THRU 37.235.
ERROR:
BLDSMPDB 37.235 CRITICAL ERROR if BLDSMPDB and VMXGALOC are used,
(only on ASCII) due to an extra comma in line 949.
"ERROR:All positional parameters must precede ..."
LIBNAME PDB NOT FOUND.
Remove that comma on line 949 in BLDSMPDB.
==MAJOR CHANGES ADDED IN MXG 37.07, DATED Oct 9, 2019 THRU 37.234.
ERROR
ASUMMIPS 37.228 Warning message DURATM=INTERVAL conflict.
ANALMSUS 37.217 EXCELDEST not protected for length 0 Warning.
CICINTRV 37.210 Warning if CICINTRV INTERVAL requested can't be used.
IMACICDB 37.200 Optional CICS DBCTL var STATCTM1 too large.
ASUMDB2A 37.199 Correct count of THREADS in DB2 ASUMDB2A/B/G/P/R.
TYPEBETA 37.198 BETA 93 Subtypes 12/17/30/31/55 are now output.
TYPEBE97 37.197 BETA 97 Subtype 51 did not input "New Area" fields.
BUILD005 37.195 PDB.PRINT only populated ACCOUNTn in first obs.
VMXGDSN 37.191 RMM/EDGR in VMXDSN had zero obs for TAPES/TAPEDSNS.
VMACEDGR 37.225 RMM/EDGR dataset EDGRXEXT RDPHYSIZE too small.
NEW SUPPORT
TYPESAPZ 37.222 Support for SAP Z Connector USER SMF record.
TYPE123A 37.221 Support for z/OS Connect EE SMF 123 Version 2.
TYPECADK 37.219 Support for CA-DISK/Sterling DMS DSINDEX file.
TYPEFOCU 37.215 Support for FOCUS Version 7.7 USER SMF Record.
TYPE80A 37.213 Support for RACF TYPE80TK TOKDANAM=AUTOLOGIN.
TYPE113 37.212 Support for z15 Processor SMF 113 RNI equation.
ASMRMFV 37.204 Support for RMF III CRYG3 and XCFG3 and more FDF.
TYPETLMS 37.192 Support for TLMS creates two new datasets.
TYPE0203 37.216 Support SMF Type 2 ST 1/2 GSIG/ISIG variable length.
ENHANCEMENT
TYPE42 37.194 TYPE42DS variables S42DSENT/DSCMT identify zEDC.
ANAL82AU 37.214 ANAL82AU combines SERV and USER obs for TYPE82AU.
VMXGALOC 37.224 Note if you added VMXGALOC to your IMACINIT.
==MAJOR CHANGES ADDED IN MXG 37.06, DATED AUG 30, 2019 THRU 37.190.
TYPEAAM 37.186 Support for IBM Tivoli Advanced Allocation SMF
DODSCRDT 37.189 Spurious INVALID VALUE FOR INPUT FUNCTION message
==MAJOR CHANGES ADDED IN MXG 37.06, DATED AUG 22, 2019 THRU 37.184.
ERROR
ASMRMFV 37.178 Possible S0C4 (37.05) or S0C7 (using FDF)
TYPE82 37.165 TYPE8201 variables SMF82ITE/CKD/LML/USR/PKD wrong.
TYPEBETA 37.160 BETA 93 610 (back level) subtype 40/49 wrong.
TYPE72GO 37.179 Variables METGOAL and PCTMETGOL were wrong.
NEW SUPPORT
TYPERMFV 37.167 z/OS 2.4 Updates for RMF MONITOR III datasets
TYPE74 37.166 z/OS 2.4 Updates for TYPE7402 dataset.
TYPE82 37.183 Support for SMF 82 new Audit TYPE82AU & subtypes.
TYPEMAR 37.181 Support for Hitachi MAR Mainframe Analytics 9.1
TYPEIMS 37.176 Support for IMS LOG TYPE '02'x.
ENHANCEMENT
TYPE1131 37.175 New SIISPCT=STORE INTO*INSTRUCTION*STREAM*PERCENT.
TYPE1415 37.172 Variable SMF14DEF='Y' if dataset is encrypted.
TYPE110 37.168 CICS "identity" variables not kept with UTILEXCL.
DODSCRDT 37.161 New z/OS-ONLY CREATDATE variable can be created.
ANALMSUS 37.157 MSU Consumption from TYPE89 and TYPE30 charts etc.
==MAJOR CHANGES ADDED IN MXG 37.05, DATED Jul 8, 2019 THRU 37.154.
ERROR
TYPE110 37.154 SMF 110 Subtype 1 MNSEGCL=5 INPUT EXCEEDED error.
==MAJOR CHANGES ADDED IN MXG 37.05, DATED Jul 6, 2019 THRU 37.153.
ERROR
TYPE120 37.153 SMF 120 Subtype 3 INPUT STATEMENT EXCEEDED error.
==MAJOR CHANGES ADDED IN MXG 37.05, DATED Jul 5, 2019 THRU 37.153.
FLASH: 37.144: MISSING PERIOD 2/3 OBS IN TYPE72GO MXG 36.07 or prior
after IBM RMF Maintenance for SCM and CRYPTO are applied.
There is NO ERROR with MXG 36.08 (Oct 2018) or later. See Text.
NEW SUPPORT
ANALMSUS 37.157 Powerful set of reports of SOFTWARE MSUs consumption.
READRATE 37.142 New analysis of Read Rate MB/Sec reading SMF data.
TYPEVM 37.130 New VM Account datasets supported.
TYPE42 37.135 Eight more invalid LENSR= TYPE42 subtype 5.
TYPE119 37.127 New variables in datasets TYP11902/TYP11994/TYP11995.
TYPE123A 37.125 Variable/format changes z/OS Connect EE 3.0 SMF 123.
ENHANCEMENT
MXGSTEP 37.152 MXGSTEP tailoring identifies MXG executions.
==MAJOR CHANGES ADDED IN MXG 37.04, DATED Jun 5, 2019 THRU 37.124.
ERRORS CORRECTED
TYPE1415 37.116 WPS 4.1 ONLY, U4087 ABEND,OPTIONS NOWPSSCATTERCOMP.
TYPENDM 37.113 NDM-CDI 24 byte short record INPUT EXCEEDED.
TYPETPX 37.107 Misaligned TPXETIME reported as 8a Oct 27, 1935.
TYPE102 37.100 DB2 zPARM T106S102 variables misaligned.
NEW SUPPORT
TYPEBETA 37.114 Support for updated BETA 93 V6R2 (INCOMPATIBLE).
TYPE7072 37.109 Support for z/OS 2.4 SMF Manual 04Mar19 changes.
TYPE120 37.105 Support for SMF 120 WAS and LIBERTY (COMPATIBLE).
TYPEIMS 37.103 Support for IMS Log Records 5607/5610/5904/5950.
TYPE110 37.102 Support for CICS/TS 5.5 new Statistics (COMPAT).
TYPE110 37.102 All _SCICxxx Statistic SORTS deaccumulate.
TYPE29 37.093 Support for IMS ODBM Accounting SMF 29 Subtype 1.
TYPETMPX 37.121 Support for ThruPut Manager Release 18.02 v7r1.0.
ENHANCEMENTS
TYPEWSF 37.111 Final revisions for WSF/EOS WSFAUDIT AUDACT/OBJN.
BLDSMPDB 37.106 Updated features and documentation.
TYPE7072 37.104 Variables CECSER/CPCMODEL added to TYPE72GO.
TYPECIMS 37.095 New variables in TYPEDBDS (IMF from BMC).
TECHNOTES
TECHNOTE 37.110 Difference between TYPExxxx and TYPSxxxx.
TECHNOTE 37.097 APAR OA65762 NEGATIVE SMF30_TIME z/OS 2.2 only.
==MAJOR CHANGES ADDED IN MXG 37.03, DATED Apr 19, 2019 THRU 37.091.
ERRORS CORRECTED
TYPE92 37.085 SMF 92 Subtype 52 INPUT EXCEEDED, TRSN doc 52 bytes.
TYPEVMXA 37.084 z/VM VXPRCAPM dataset vars CMB10C0-X4 wrong values.
TYPEBE97 37.080A Datasets BETA9706/BETA9706D were not output to PDB.
TYPE74 37.078 TYPE748S var R748SIID fmt $HEX4, no dupes now.
TYPE125 37.075 INPUT STATEMENT EXCEEDED, period missing.
TYPEDB2 37.074 QBSTBPIN always incorrectly calculated before DIF.
VMAC82 37.060 INPUT EXCEEDED SMF 82 ST 31, incorrect length.
NEW SUPPORT
TYPERMFV 37.067 Support for RMF III PCI/SCM/ZFX segments 4 datasets.
TYPEHSM 37.076 Support for HSM FSR Record, Unix filename added.
TYPE99 37.082 SMF 99 ST 12 Capacity Incr/Decr individual decodes.
ANALRMF3 37.068 CF Activity Report, Structure Level, in ANALRMF3.
TYPEDCOL 37.069 zEDC Compression type values revised DCOLDSET/DCOLDC.
TYPE102 37.059 Final corrections for IFCIC 319 support.
ENHANCEMENTS
TYPEXAM 37.081 Analyzing VPS USER dataset, must use INTORSUM='SU'.
TYPEWSF 37.083 Logic revised to use OBJT/ACT for Input choice.
TYPERMFV 37.080 SVPCNM and RPRTCLAS added to all RCD datasets.
TYPEBBMQ 37.073 UNEXPECTED RTIN messages, BBMQ Version 5.4 no change.
ANALID 37.063 Report now shows 26.002 or 26.003 for JES2/JES3.
GRAFWLM 37.061 Bar charts of ZIP and ZIP Eligible added.
TYPE30 37.058 Cosmetic: Uninitialized variable CBPERROR no impact.
TECHNOTE 37.072 ODS Stat graphics procs use JAVA, memory intensive.
==MAJOR CHANGES ADDED IN MXG 37.02, DATED Mar 11, 2019 THRU 37.057.
ERRORS CORRECTED
TYPENDM 37.047 NDM-CDI dataset NDMCT var NDMCPU 256 times too large.
READDB2 37.042 MXG 37.01. Blank WANTONLY Cosmetic %SYSFUNC message
TYPE74 37.040 TYPE749 variable R7491DEFCOMPRATIO wrong value.
TYPE42 37.034 Two more ABENDS invalid SMF 42 LENSR 520/592 protect.
TYPE74 37.032 TYPE749 z/EDC Divide by ZERO protection failed.
TYPE7072 37.044 BMC CMF VERSNRMF values 792 and 794 for z/OS 2.3.
NEW SUPPORT
TYPEDCOL 37.041 Support for APAR OA54897, DCDEXFLG not used FOR zEDC.
MANY 37.037 Support for SMF Manual Changes in Jan 14, 2019 Doc.
TYPEAXW 37.033 Support for Axway V3.3.2 2018/06/27 restructure.
UTILMISS 37.053 Utility to remove all variables that are all missing.
TYPEDB2 37.035 DB2 V12 overlooked Package variables in DB2ACCTP.
TYPE102 37.051 IFCID 319 new variables created and kept.
ENHANCEMENT
JCLPDBJB 37.048 Example "BUIDPDB" creates only JOB-related datasets.
TYPERMFV 37.055 CFACT Coupling Facility Structure Activity Report.
ANALMQ 37.039 MQ Reports replicating IBM's MQSMF program.
ANALHSM1 37.038 Combined TYPE6156+HSMFSRST report, thrashing pri-mig?
TYPE70PR 37.046 SMF70BPS/SMF70ACS expanded for each engine type.
TECHNOTE 37.043 Executing MXG on ASCII, WORK needs to be local.
==MAJOR CHANGES ADDED IN MXG 37.01, DATED Feb 3, 2019 THRU 37.030.
ASMRMFV 37.030 ASMRMFV 36.12-37.01 NOZEROCPU filter didn't.
Caused out of space condition. Typically, NOZEROCPU
filters out 3/4 of the ZRBASI records.
==MAJOR CHANGES ADDED IN MXG 37.01, DATED Feb 1, 2019 THRU 37.030.
ABEND Avoidance
TYPEDB2 37:014 ABEND: DB2 SMF 100 ST 1 NETEAZZA/IDAA DB2STATS.
TYPEVMXA 37.012 ABEND: ZVM MONWRITE NEW 40061802 Service Level.
TYPEVMXA 37.028 ABEND: ZVM MONWRITE new z/VM 7.1 (INCOMPAT).
ERRORS CORRECTED
TYPE92 37.002 TYPE9208 INPUT STATEMENT EXCEEDED, manual wrong.
TYPERMFV 37.001 Some RMF III ZRBASI fields blank/wrong in 36.36.
TYPE119 37.003 TYP11952 SMF119ML_IP_IPV4 wrongly compressed TIRIP.
VMXGPRAL 37.006 Unbalanced parens in variable label, non fatal.
TYPE92 37.017 Many non-fatal corrections were made to type 92.
TYPEEDGR 37.015 RMM variable EDGRTIME had missing values.
ANALCNCR 37.013 New Concurrency example counts steps and tapes.
NEW SUPPORT
TYPEBETA 37.007 Support for Beta93 V6.2 Subtypes 1-3.
TYPE102 37.005 Support for DB2 102 Trace IFCID 404.
TYPESTC 37.018 Support for STC HSC Subtype 32 and 33 new datasets.
TYPE26J2 37.026 Local SubSystem TYPE26J2 not output, SMF6SBS NE 2.
VMXGALOC 37.021 New parms YR2KEEP and BASEYEAR for Yearly PDB.
ENHANCEMENT
EMAIL 37.027 Example added to email SAS CondCode from ASCII SAS.
ANALID 37.016 New report showing total/min/max for each SYSTEM.
VMXGUOW 37.011 Enhanced for easy CICSTRAN-only PDB.ASUMUOW.
ANAL89 37.029 Reference line SMF70LAC (4HRAV) added to MSU plots.
TECHNOTES
TECHNOTE 37.004 Reading z/OS DATA with SAS FTP Access needs RCMD
Please read CHANGESS for the complete list of major enhancements.
See member NEWSLTRS or the Newsletters frame at http://www.mxg.com for
current MXG Technical Notes.
All of these enhancements are described in the Change Log, below.
II. SAS Version requirement information:
SAS Versions
The current version nomenclature is SAS 9.4 TS1M6 (9.4M6), "M6",
or "SAS 9.4 (9.04.01M6P110718)" if the OPTION VERSIONLONG is
enabled.
Only on z/OS, SAS 9.4 "M5" requires MXG 35.36+ because it adds the
NOERRORSTOP option to protect all MXG PROC SQLs from the M5 defect
described in SAS Note 61672. But SAS apparently does not plan for
a defect correction since the MXG Circumvention solves for MXG and
the text of 61672 simply describes the circumvention needed because
MXG's use of OPTIONS OBS=0 without NOERRORSTOP exposed the defect.
See Change 35.309 for more details on using NOERRORSTOP for your
own PROC SQLs.
SAS V9.4 M6 is RECOMMENDED, but MXG executes without error
using SAS Version 9.4 M0-M2 or M4-M6 or SAS Version 9.3 M0-M2.
SAS V9.4 M5 is REQUIRED with z/OS 2.3 with Eight-Byte USERIDs
for Interactive TSO (DMS) SAS Sessions. SAS Note 61339.
SAS V9.4 M3 is NOT RECOMMENDED. See Change 36.128 SAS Note 61906
that reports 40% Increase in CPU time with M3.
SAS V9.4 (ALL) and SAS V9.3 (ALL) are at LEVEL A SAS Support.
SAS V9.3 SAS 9.3 TS1M2 was RECOMMENDED. SAS 9.3 TS1M1 works ok.
But SAS 9.3 at TS1M0, the HOT FIX for SAS Note SN-43828,
see CHANGE 29.169, IS REQUIRED:
The %MACRO compiler error is in processing %LET
statements. While only two MXG members failed
repeatedly in MXG QA tests on z/OS, there were random
%LET errors in ASCII QA tests, so ANY use of %LET
statement on ANY platform are vulnerable to this
error, as the %MACRO compiler is SAS portable code,
used on all platforms. So this is NOT just an MXG
error, but impacts ALL SAS programs.
SAS9.3 is LEVEL A support from SAS.
SAS V9.2 Was recommended, prior to 9.3, and was error-free with
MXG 26.03 SAS Hot Fix for SAS Note 37166 is required to
use a VIEW with the MXG EXITCICS/CICSFIUE CICS/DB2
Decompression Infile Exit. but SAS V9.2 does execute on
that platform.
9.2 is LEVEL B Support from SAS, as of Sep 30, 2013.
SAS V9.1.3 on z/OS 1.10 requires SAS Hot Fix for SN-35332 and is at
Support level C by SAS Institute, Sep 30, 2013.
SAS V9.1.3 is NOT supported by SAS on Windows SEVEN.
SAS V8.2 SUPPORT LEVEL C BY SAS INSTITUTE; NOT ALL OF MXG WORKS!
with SAS 8.2.
SAS 8.2 is Level C Support from SAS as of Dec 31, 2011.
JCL in MXGSAS94 or MXGSAS93 can be used, or MXGNAMES can be used
***************************************************************
As documented in Change 27.356, for SAS V9.2 or later):
The standard SAS JCL Procedure can be used for MXG with SAS V9.2+
// EXEC SAS,CONFIG='MXG.SOURCLIB(CONFIMXG)'
//MXGNAMES DD DSN=MXG.USERID.SOURCLIB(MXGNAMES),DISP=SHR
But CONFIMXG is required for sites with NLS issues, and you must
use JCLCONFI to create/update the MXG.FORMATS catalog if you use
CONFIG='MXG.SOURCLIB(CONFIMXG)'.
For no NLS, you can use the MXGSAS94 JCL Procedure example.
***************************************************************
MXG 26.03 thru MXG 36.11 will execute under the previously listed
SAS Versions on all supported platforms
Unrelated to the above SAS Note/Hot Fix, ODS users will want to
use MXG 29.06+, because SAS V9.3 did expose incompatibilities in
MXG code for ODS reporting, that were fixed in MXG Version 29.06.
See Changes 29.159 and 29.169.
And, only for z/OS 1.10 with SAS V9.1.3 with ANY version of MXG,
the SAS Hot Fix for SN-35332 is REQUIRED (to be completely safe).
Without this Hot Fix, "LIBREF XXXXXXXX IS NOT ASSIGNED" errors
can occur even though //XXXXXXXX DD is a valid SAS Data Library.
This error ONLY occurs with z/OS 1.10 and SAS V9.1.3; it does
NOT occur with SAS V9.2, nor with z/OS 1.9. It can be
circumvented by adding a LIBNAME statement that specifies the
ENGINE name. See the Technical Note in Newsletters for SN-35332.
Note that SAS V9.1.3 is now at "Level B" Support from SAS.
Old MXG code may continue to execute with SAS V8.2, but V8 is now
"Level C" support from SAS Institute, and there are known errors
in V8.2 that are only fixed in SAS V9. I no longer QA with V8.2;
While many MXG programs (accidentally) will still execute under
V8.2, I cannot guarantee that all of MXG executes error free.
PLEASE INSTALL V9.2/V9.3/V9.4, TO AVOID FIXED PROBLEMS!
If you are absolutely stuck on V8, you need to copy MXG member
V8GETOBS into USERID.SOURCLIB and rename to VGETOBS.
MXG Software has not executed under SAS V6 in many years.
The "PDB" libraries (i.e., SAS data libraries) must be created by
SAS V8 or later, but any of those data libraries can be read or
updated by the SAS Versions that MXG Supports, above.
For SAS Version V9.3:
SAS 93 TS1M1 is RECOMMENDED; for TS1M0, SAS Hot Fix in SAS Note
SN43828 is REQUIRED. See text of Change 29.159.
With SAS 93 TS1M1, (or TS1M0 with that Hot Fix) MXG Versions
26.03 or later execute under SAS V9.3 on all platforms.
SAS Data Libraries created by SAS V8.2, V9.1.3, V9.2, V9.3 and
SAS V9.4 are interchangeable and can be read/written by any of
those versions, provided they are on the same platform.
BUT: on ASCII, the 32-bit and 64-bit SAS versions are NOT the
same "platform" and attempting to read/use the FORMAT catalog
created on one of those "platforms" on the other "platform"
will error out to remind you of that difference!
SAS V9.4 did change some V9.3 ODS processing defaults and syntax
that might cause errors with MXG 29.05 or earlier; MXG 29.06,
Change 29.160 documents the major revisions made in MXG to fully
support ODS, and MXG 29.06 is STRONGLY recommended for ODS with
SAS V9.3 or SAS V9.4.
For (Archaic) SAS Version V9.2 (TS1M0):
Big Picture: SAS Version V9.2 is COMPATIBLE with MXG Software.
On z/OS, SAS changed the DSNAMES for some of the SAS libraries,
so you do need to use the new MXGSAS92 JCL Procedure for MXG,
but it still uses the CONFIGV9 configuration file.
****************************************************************
However, NEW, and documented in Change 27.356, with SAS V9.2+:
The standard SAS JCL Procedure can be used for MXG:
// EXEC SAS,CONFIG='MXG.SOURCLIB(CONFIMXG)'
//MXGNAMES DD DSN=MXG.USERID.SOURCLIB(MXGNAMES),DISP=SHR
instead of using the MXGSAS92 JCL Procedure example.
****************************************************************
SAS Data Libraries are compatible for V8.2, V9.1.3, V9.2, V9.3,
and V9.4. "PDBs" can be read/written interchangeably between
these SAS versions.
MXG Versions 26.03+ do execute with SAS V9.2 with NO WARNINGS
and with NO ERRORS reported.
Pre-MXG 26.03, SAS Hot Fix F9BA07 was required to suppress a
new SAS V9.2 WARNING, that on z/OS, set CC=4 (condition/return
code). That warning is harmless (to MXG code) and all MXG
created SAS datasets were correct, even with that warning.
The ONLY exposure was ONLY on z/OS, and ONLY if condition code
tests are used in your MXG jobstreams.
SAS Version 9.2 requires z/OS 1.7 or later, both officially as
documented by SAS Institute, and actually as V9.2 fails with 0C4
under z/OS 1.4.
For SAS V9.1.3 on z/OS with Service Pack 4:
On z/OS 1.10, Hot Fix SN-35332 is REQUIRED.
CONFIGV9 now specifies V9SEQ instead of V6SEQ. As V6SEQ does
not support long length character variables, it can't be used.
SAS V9.1.3 with current Service Pack 4 is STRONGLY RECOMMENDED.
For (back-level!) SAS V9.1 or V9.1.2 on z/OS:
SN-013514 is REQUIRED to be able to read datasets that were
created by V6SEQ (tape) engine.
SN-012437 is REQUIRED to prevent creation of corrupt/unreadable
datasets with tape engines V7SEQ, V8SEQ, or V9SEQ.
Both fixes ARE included in SAS V9.1.3, but V9.1 or 9.1.2 is NOT
SAFE without those two hot fixes, and if you do NOT have those
two fixes on 9.1 or 9.1.2, you MUST set V6SEQ in CONFIGV9.
With MXG 23.02 or later, V9SEQ is the default sequential engine
specified in CONFIGV9, but if you are back at SAS V9.1 or V9.1.2
you MUST install the two hot fixes listed above.
For SAS Version 8.2, HotFix Bundle 82BX08 (or later) was required
as an absolute minimum level when that SAS Version was last
supported by MXG Software. PLEASE INSTALL SAS V9.x ASAP.
Sequential Engine Status:
V9SEQ was fixed in V9.1.3; it has been default in CONFIGV9.
V8SEQ was always safe under SAS V8.2, but it wasted CPU time
by always compressing when writing in tape format.
V6SEQ, if used under V9.1.2, requires SN-013514, but V6SEQ
should no longer be used, as it does not support long
length variables.
GENERAL STATEMENT FOR MXG QA TESTS AND SAS VERSIONS:
MXG QA tests are executed with V9.4, on z/OS, on Windows TEN and
Linux on 64-bit hardware, but MXG users execute MXG on MANY
(ALL??) SAS platforms, including AIX, Linux, and other 'nix'
variants, on many different hardware platforms, and since they all
work we don't need to list them. If SAS executes so does MXG.
Prior QA tests have been run with all SAS releases available at
that time on Linux RH8 on Intel, on Solaris v2.8 on a Model V880,
and on HP-UX v11.11 model rp5470, confirming full compatibility.
MXG should execute under ALL SUPPORTED SAS VERSIONS on EVERY SAS
platform without errors! Each new MXG version is also tested with
the SAS ITSV/ITRM product by the ITRM developers.
III. WPS Version requirement information:
WPS Version 3.02 (03.02.03.00.016221) is required Change 34.266.
and other errors with 3.00 or 3.01 have been corrected in the
current WPS version.
WPS Version 3.01.1 maintenance level 731 required for PDB to tape
WPS Version 3.01 (also shows 3.1.1) is required for AUTOEZOS.
WPS Version 3.01 is required for MOBILWRK, PICTURE fails in 2.5.
WPS Version 3.01 executed MXG 32.03 BUILDPDB with no errors.
WPS Version 3.0 requires MXG 31.09 (see Change 31.251).
WPS Version 2.4 required MXG 27.09 (see Change 27.239).
WPS Version 2.3.5 required MXG 27.05.
See NEWSLETTERS for WPS Support Statement.
WPS prints this message ERROR: COULD NOT CREATE DATA SET "PDB.ID"
when the LIBNAME PDB does not exist; there would also have been a
prior log message NOTE: Library PDB does not exist as the clue.
IV. MXG Version Required for Hardware, Operating System Release, etc.
MXG is usually NOT sensitive to z/OS Hardware changes, but:
The z15 did INCOMPATIBLY change the SMF 113 records by adding 32
new EXTEND and 4 CRYPTO counters, and those fields were supported
in MXG 37.07 dated Oct 22, 2019. The z/14 also INCOMPATIBLY changed
the SMF 113 record, and that was supported way back in MXG 36.07.
The z/13 with 61+ LPARs requires MXG 32.05 IF NON-SMT MODE.
The z/EC12 with 85+ engines required MXG 30.07.
Support for 255 engines was added in MXG 31.04.
MXG 37.07 supports the new z/OS 2.4 SMF manual fields, COMPATIBLY.
The z13 processor INCOMPATIBLY CHANGED, the new SMT-MODE RMF 70, and
MXG 34.03 was REQUIRED (PCTCPUBY WRONG!), to read the SMT-format RMF
(which are written if you have zIIP engines AND have enabled the new
PROCVIEW CORE option for Multi-Threading, even if only one thread is
enabled).
The new zEDC compression hardware requires MXG 33.07 to support the
new metrics.
For z/VM, MXG REQUIRES MXG 33.02 to support the z/13 changes.
Availability dates for the IBM products and MXG version required for
error-free processing of that product's data records:
Product's
Availability MXG Version
Product Name Date Required
MVS/ESA 4.1 Oct 26, 1990 8.8
MVS/ESA 4.2 Mar 29, 1991 9.9
MVS/ESA 4.2.2 Aug 15, 1991 9.9
MVS/ESA 4.3 Mar 23, 1993 10.10
MVS/ESA 5.1.0 - compatibility Jun 24, 1994 12.02
MVS/ESA 5.1.0 - Goal Mode May 3, 1995 13.01
MVS/ESA 5.2.0 Jun 15, 1995 13.05
MVS/ESA 5.2.2 Oct 19, 1995 13.09
OS/390 1.1.0 Feb 22, 1996 14.01
OS/390 1.2.0 Sep 30, 1996 14.05
OS/390 1.3.0 Compatibility Mode Mar 28, 1997 14.14
OS/390 1.3.0 WLM Goal Mode Mar 28, 1997 15.02
OS/390 2.4.0 Sep 28, 1997 15.06
OS/390 2.5.0 Feb 24, 1998 15.06
OS/390 2.6.0 Sep 24, 1998 16.04
OS/390 2.7.0 Mar 26, 1999 16.09
OS/390 2.7.0 APAR OW41318 Mar 31, 2000 18.03
OS/390 2.8.0 Aug 24, 1999 16.09
OS/390 2.8.0 FICON/SHARK Aug 24, 1999 17.08
OS/390 2.8.0 APAR OW41317 Mar 31, 2000 18.03
OS/390 2.9.0 Mar 31, 2000 18.03
OS/390 2.10.0 Sep 15, 2000 18.06
OS/390 PAV Oct 24, 2000 18.09
z/OS 1.1 Mar 30, 2001 18.11
z/OS 1.1 on 2064s Mar 30, 2001 19.01
z/OS 1.1 with correct MSU Mar 30, 2001 19.02
z/OS 1.2 Oct 31, 2001 19.04
z/OS 1.1,1.2 APARs to 78 Oct 31, 2001 19.05
z/OS 1.2+ APAR OW52227 Apr 26, 2002 20.02
z/OS 1.3+ APAR OW52227 Apr 26, 2002 20.02
z/OS 1.2 JESNR Z2 MODE Apr 26, 2002 20.03
z/OS 1.3 JESNR Z2 MODE Apr 26, 2002 20.03
z/OS 1.4 Tolerate Sep 27, 2002 20.03
z/OS 1.4 Support Sep 27, 2002 20.06
z/OS 1.4 Over 16 CPUs/LPARs May 29, 2003 21.02
z/OS 1.4 DFSMS/rmm, RACF Aug 29, 2003 21.04
z/OS 1.5 Mar 31, 2004 21.21
z/OS IRD ASUM70PR/ASUMCEC Sep 22, 2003 *24.10
z/OS IRD TYPE70PR Mar 11, 2004 *24.10
z/OS IRD TYPE70,RMFINTRV Mar 22, 2002 *24.10
z/OS 1.6 - No IFAs Sep 30, 2004 *22.09
z/OS 1.6 - With IFAs Sep 30, 2004 *22.11
z/OS 1.7 (COMPATIBLE CHANGES) Sep 30, 2005 *24.10
z/OS 1.7 (SPLIT70 CORRECTION) Sep 30, 2005 *24.10
z/OS IFA data in RMF 79s Sep 30, 2005 23.10
z/OS 1.8 - ASMTAPEE assembly Sep 30, 2005 *25.03
z/OS 1.8 - SMF 119 INCOMPAT Sep 30, 2005 *25.06
z/OS More than 32 LPARs Jan 30, 2006 *24.24
z/OS SPLIT RMF 70 records Jan 30, 2006 *24.24
z/OS Dupe SYSTEMs in a SYSPLEX Jan 30, 2006 *24.02
z/OS IRD errors corrected May 15, 2006 24.03
z/OS ASUMCEC errors corrected May 15, 2006 *24.24
z/OS ASUM70LP errors corrected Jun 13, 2006 *24.24
z/OS zIIP Processor Support Jun 22, 2006 *24.24
z/OS Dedicated zIIP Support Mar 8, 2008 *26.01
z/OS Dedicated zAAP Support Mar 8, 2008 26.01
z/OS 1.8 (COMPATIBLE CHANGES) Sep 20, 2006 *24.24
z/OS 1.9 (INCOMPAT, 54 CPs) Sep 27, 2007 25.10
z/OS 1.9 MXGTMNT at ML-39 reASM Sep 27, 2007 25.10
z/OS new z10 variables Mar 5, 2008 26.01
z/OS 1.8 With HiperDispatch Sep 15, 2008 *26.10
z/OS 1.9 With HiperDispatch Sep 15, 2008 *26.10
z/OS 1.10 (INCOMPAT, MXG code) Sep 15, 2008 26.07
z/OS 1.10 With HiperDispatch Sep 15, 2008 *26.10
z/OS 1.10 RMF III, SMF 119 Jul 20, 2009 27.05
z/OS 1.11 Sep 2, 2009 27.08
z/OS 1.11 New 30 variables Apr 14, 2010 *28.02
z/OS 1.12 Aug 17, 2010 *28.05
z/OS 1.12 SMF 85 Subtype 79 Aug 17, 2010 *29.03
z/OS 1.12 VMGUEST option Aug 17, 2010 *29.06
z/OS 1.13 Sep 30, 2011 29.03
z/OS 1.13 - MXGTMNT only Dec 15, 2011 29.08
z/OS 1.13 SMF 119 ST 6 INCOMPAT Feb 7, 2012 30.01
z/OS 2.1 - Most Records support Jul 23, 2013 30.05
z/OS 2.1 - ID=0 ERROR MESSAGE Jul 23, 2013 31.07
z/OS 2.1 - ID=85 INCOMPAT Jul 23, 2013 32.03
z/OS 2.1 - ID=70 SMF70CPA Jul 23, 2013 32.03
z/OS 2.1 - INPUT STATEMENT EXCEEDED ERROR SMF 74 33.10
z/OS 2.2 COMPATIBLE CH 33.189 Aug 19, 2015 33.08
z/OS 2.2 MXGTMNT ABEND S0E0-28 Sep 15, 2015 33.09
REQUIRES ASMTAPE ML-55 Sep 15, 2015 33.09
z/OS 2.2 OAM SMF 85 ABEND 33.067 Apr 5, 2016 34.02
z/OS 2.2 SPLIT 73, ABEND 33.068 Apr 5, 2016 34.02
z/OS 2.2 JES2 8-char JOBCLASS Oct 7, 2016 34.07
z/OS 2.2 NEW SMF 124 IOS Spvr Oct 7, 2016 34.07
z/OS 2.3 Many new variables Sep 24, 2017 35.166 35.09*
z/OS 2.3 RMF III Support Sep 24, 2017 35.190 35.09*
z/OS 2.3 type 2 st 2 STOPOVER Sep 24, 2017 35.190 35.09*
z/OS 2.3 type 90 st 38 STOPOVER Sep 24, 2017 35.199 35.09*
z/OS 2.4 Compatible from SMF Manual Sep 2019 37.166 37.07.
z990 CPUs - CPUTYPE '2084'x Aug 25, 2003 21.04
z890 CPUs - CPUTYPE '2086'x Jun 24, 2004 22.07
z9 CPUs - CPUTYPE '2094'x Jul 20, 2005 *24.24
z9EC CPUs - CPUTYPE '2094'x:
with 64-bit z/OS - no change required *24.24
with 32-bit z/OS only: Aug 26, 2006 24.06
z9BC CPUs - CPUTYPE '2096'x:
with 64-bit z/OS - no change required 24.01
with 32-bit z/OS only: Jul 27, 2006 *24.24
z10 CPUs - CPUTYPE '2097'x Dec 7, 2008 25.11
z10 HiperDispatch/Parked Time Mar 3, 2008 *26.10
z196 (INCOMPAT IF GT 64 ENG) Aug 17, 2010 28.05
zEC12 Nov 14, 2012 30.07
z13 non-SMT Mode May 27, 2014 32.05
z13 SMT Mode Change 33.217 Sep 15, 2015 *33.09
z13 SMT Mode NRZIPCPU 34.106 May 10, 2016 34.03
z13 SMT MT=2 CPUZIPTM TYPE70 Mar 21, 2016 35.03
z14 SMF 113 INCOMPAT, ABEND Oct 2, 2017 35.11
z14 113 LPARBUSY missing value Aug 8, 2018 36.07
z14 ZR1 New SMF70MAXPU variable May 8, 2018 36.04
z15 New SMF fields (COMPAT) Oct 22, 2018 37.07
CICS/CTG V9 Transaction Gateway ?? ?? 2013 31.31
CICS/ESA 3.2 Jun 28, 1991 9.9
CICS/ESA 3.3 Mar 28, 1992 10.01
CICS/ESA 4.1 Oct 27, 1994 13.09
CICS/ESA 5.1 aka CICS/TS V1R1 Sep 10, 1996 14.07
CICS-Transaction Server V1R1 Sep 10, 1996 14.07
CICS-TS V1R1 with APAR UN98309 Sep 15, 1997 15.06
CICS-TS V1R2 CICS/TS 1.2 Oct 27, 1997 15.06
CICS-TS V1R3 CICS/TS 1.3 Mar 15, 1999 17.04
CICS-TS V2R1 CICS/TS 2.1 Mar 15, 2001 18.11
CICS-TS V2R2 CICS/TS 2.2 Feb 9, 2002 19.19
CICSTRAN subtype 1 support only *19.19
CICSTRAN subtype 2 completed *19.08
CICS-TS V2R3 CICS?TS 2.3 Dec 19, 2003
Using UTILEXCL to create IMACEXCL: 21.04
Reading un-Excluded CICS with TYPE110, no IMACEXCL:*22.04
CICS-TS V2R3 CICS/TS 2.3 Aug 13, 2004 22.04
CICS-TS V3R1 CICS/TS 3.1 Jan 18, 2005 22.22
CICS-TS for Z/OS Version 3.1 Mar 15, 2005
Using UTILEXCL to create IMACEXCL: 22.13
Reading un-Excluded CICS with TYPE110, no IMACEXCL: 22.22
CICS-TS V3R2 CICS/TS 3.2 Dec 6, 2007 25.11
CICS-TS for Z/OS Version 3.2 Jun 29, 2007 25.03
CICS-TS/3.2 Compressed Records Nov 3, 2007 25.11
CICS-TS/4.1 (CICSTRAN INCOMPAT) Mar 13, 2009 27.01
CICS-TS/4.1 (STATISTICS ST=2) Sep 18, 2009 27.08
CICS-TS/4.2 CICSTRAN/STATISTICS Jun 24, 2011 29.03
CICS-TS/4.2 CICSRDS MNSEGCL=5 Jun 24, 2011 *29.05
CICS-TS/4.2 INVALID STID=116 Jan 31, 2012 *30.01
CICS-TS/5.1 (INCOMPATIBLE) Dec 14, 2012 *30.08
CICS-TS/5.1 for valid TASZIP/ELG Jan 21, 2013 *30.30
CICS-TS/5.1 MNSEGCL=5 INCOMPAT Jun 17, 2013 *31.03
CICS-TS/5.2 COMPATIBLE CICSTRAN Jun 13, 2014 *31.03
CICS-TS/5.2 INCOMPAT Statistics Jun 13, 2014 *32.03
CICS-TS/5.3 INCOMPAT CICSTRAN Apr 29, 2015 33.04
CICS-TS/5.3 RESOURCE SEGCL=5 Sep 31, 2015 33.09
CICS-TS/5.3 CICSTRAN INCOMPATIBL Oct 29, 2015 33.11
CICS-TS/5.3 GA date Dec 11, 2015 33.33
CICS-TS/5.3 MNSEGCL=5 INPUT ERR Mar 21, 2016 34.02
CICS-TS/5.4 OPEN BETA Aug Aug 11, 2016 34.06
CICS-TS/5.4 OPEN BETA Nov Nov 11, 2016 34.09
CICS-TS/5.4 GA Jun 17, 2017 35.03
CICS-TS/5.5 GA (COMPAT) Jan 29, 2018 37.01
DB2 2.3.0 Oct 28, 1991 10.01
DB2 3.1.0 Dec 17, 1993 13.02
DB2 4.1.0 Tolerate Nov 7, 1995 13.07
DB2 4.1.0 Full support Sep 11, 1996 14.07
DB2 5.1.0 Tolerate Jun 27, 1997 14.14
DB2 5.1.0 Full support Jun 27, 1997 15.02
DB2 6.1.0 initial support Mar 15, 1999 16.09
DB2 6.1.0 all buffer pools Mar 15, 1999 18.01
DB2 6.1.0 parallel DB2 Mar 15, 1999 19.19
DB2 7.1.0 parallel DB2 Mar 31, 2001 19.19
DB2 7.1.0 corrections Mar 31, 2001 20.06
DB2 8.1 Tolerate, no packages Mar 31, 2004 20.20
DB2 8.1 New Data Packages wrong Mar 31, 2004 21.08
DB2 8.1 Support with Packages Mar 31, 2004 *23.09
DB2 8.1 with all zIIP Variables Sep 30, 2006 24.08
DB2 8.1 +PK47659 Sep 12, 2008 26.08
DB2 9.1 See Change 25.265. Dec 7, 2007 25.11
DB2 9.1 Full Support +PK/56356 Sep 12, 2008 26.08
DB2 10.1 Tolerate Oct 1, 2010 28.06
DB2 10.1 New vars + Compressed Nov 1, 2010 *28.07
DB2 10.1 Invalid Header pre APAR Jan 12, 2011 *28.28
DB2 10.1 IFCID=225 INCOMPAT Sep 23, 2011 *29.07
DB2 10.1 QWHCCV for QWHCATYP=8 Oct 3, 2011 *30.07
DB2 10.1 DBID/OBID decode Jan 21, 2013 *30.30
DB2 10.1 QLSTxxxx vars corrected Jun 21, 2013 *31.04
(ONLY IMPACTS DB2STATS)
DB2 11.1 TOLERATE DB2 V11.1 Jun 21, 2013 30.30
DB2 11.1 DB2STATS QLST CORRECT Jun 21, 2013 31.04
DB2 11.1 SUPPORT NEW VARIABLES Jun 21, 2013 31.08
DB2 11.1 IRLM NEW SEGMENT Jun 21, 2013 32.10
DB2 12.1 COMPATIBLE Oct 5, 2016 34.08
DB2 12.1 NETEZZA CORRECTIONS Oct 5, 2016 34.08
DB2 12.1 QLAC INSERTS DB2ACCT May 15, 2017 35.05*
DFSMS/MVS 1.1 Mar 13, 1993 11.11
DFSMS/MVS 1.2 Jun 24, 1994 12.02
DFSMS/MVS 1.3 Dec 29, 1995 13.09
DFSMS/MVS 1.4 Sep 28, 1997 15.04
DFSMS/MVS 1.4 HSM Sep 23, 1998 16.04
DFSMS/MVS 1.5 ??? ??, 1999 16.04
DFSORT SMF V1R5 Mar 1, 2006 24.02
MQM 1.1.2, 1.1.3, 1.1.4 Apr 25, 1996 14.02
MQ Series 1.2.0 May 26, 1998 16.02
MQ Series 2.1.0 Oct 2, 1999 17.07
MQ Series 5.2 Dec 16, 2000 18.10
MQ Series 5.3 Dec 16, 2002 21.05
MQ Series 6.0 Feb 14, 2006 23.23
Websphere MQ Series 7.0 ??? ??, 2009 *28.06
Websphere MQ Series 7.1 MAR 12, 2011 29.03
Websphere MQ Series 8.0 Jun 24, 2011 29.05
Websphere MQ Series 9.1 Mar 20, 2017 35.03
NETVIEW 3.1 type 37 ??? ??, 1996 14.03
NPM 2.0 Dec 17, 1993 12.03
NPM 2.2 Aug 29, 1994 12.05
NPM 2.3 ??? ??, 1996 15.08
NPM 2.4 Nov 18, 1998 17.01
NPM 2.5 Feb ??, 2000 18.02
NPM 2.6 Nov ??, 2001 19.06
RMDS 2.1, 2.2 Dec 12, 1995 12.12
RMDS 2.3 Jan 31, 2002 19.11
TCP/IP 3.1 Jun 12, 1995 12.12
TCP/IP 3.4 Sep 22, 1998 16.04
WebSphere 5.0 APAR PQ7463 Aug 19, 2003 21.04
WebSphere 6.0 Feb 18, 2006 23.23
WebSphere 7.0 Oct 7, 2010 28.06
WebSphere 8.0 Jul 17, 2011 29.05
DOS/VSE POWER V6.3.0 Dec 19, 1998 16.08
VM/ESA 2.0 Dec 23, 1992 10.04
VM/ESA 2.1 Jun 27, 1993 12.02
VM/ESA 2.2 Nov 22, 1994 12.06
VM/ESA 2.3 Jun 1, 1998 16.08
VM/ESA 2.4 Mar 1, 2001 19.03
z/VM 3.1 Mar 1, 2001 19.03
z/VM 3.1 DATABYTE=0 May 2, 2002 20.02
z/VM 4.2 ?? May 2, 2002 20.02
z/VM 4.4 Jan 22, 2005 22.22
z/VM 5.1 Jan 22, 2005 22.22
z/VM 5.2 Jan 22, 2006 24.01
z/VM 5.3 TOLERATE Jun 7, 2007 25.05
z/VM 5.3 NEW VARIABLES Sep 12, 2008 26.08
z/VM 5.4 (COMPATIBLE) Sep 12, 2008 *27.01
z/VM 6.1 (NO CHANGES) Jul 7, 2008 27.01
z/VM 6.2 Dec 2, 2011 29.04
z/VM 6.3 INCOMPATIBLE Jul 23, 2013 31.05
z/VM 6.3 z/13 Jan 23, 2016 33.33
z/VM 6.4 SYTLCK Incompat Apr 26, 2016 34.04
z/VM 6.40061802 ABEND Jan 17, 2019 37.02
z/VM 7.1 ABEND Feb 14, 2019 37.02
IMS log 4.1 Jul 4, 1994 12.02
IMS log 5.1 Jun 9, 1996 14.05
IMS log 6.1 ??? ?, 199? 20.03
IMS log 7.1 ??? ?, 200? 20.03
IMS log 8.1 May 21, 2003 21.02
IMS log 9.1 Mar 96, 2004 *26.01
IMS log 10.1 Mar 06, 2007 *26.01
IMS log 11.1 Apr 1, 2010 *28.02
IMS log 12.1 Jan 23, 2012 *29.29
IMS log 13.1 (NOT 56FA) May 25, 2013 31.03
IMS log 13.1 (56FA RECORD) May 27, 2014 32.05
IMS log 14.1 COMPATIBLE Dec 19, 2015 33.07
IMS log 15.1 NO CHANGES Mar 1, 2018 35.07
AS400 3.7.0 Nov 1, 1996 15.01
AS400 4.1.0 Dec 30, 1996 15.08
AS400 4.2.0 Apr 27, 1998 16.02
AS400 4.4.0 Sep 27, 1999 17.07
AS400 4.5.0 Jul 27, 2000 18.07
AS400 5.2.0 - Most records Jul 23, 2003 21.03
AS400 5.2.0 - QAPMMIOP Jul 23, 2003 22.04
AS400 5.3.0 Jan 22, 2005 22.22
AS400 5.4.0 Aug 26, 2006 24.06
AS400 6.1.0 Jun 29, 2008 26.05
Note: Asterisk by the version number means the Version number
was changed (to the MXG version required), after an earlier
MXG version was listed as supporting this product release,
usually because an APAR modified the product's data records.
Or a coding error in MXG could be the reason for the change!
Availability dates for non-IBM products and MXG version required:
MXG Version
Product Name Required
Demand Technology
NTSMF Version 1 Beta 14.11
NTSMF Version 2.0 15.05
NTSMF Version 2.1 15.06
NTSMF Version 2.2 16.04
NTSMF Version 2.3 17.10
NTSMF 2.4.4 Aug 9, 2002 20.04
NTSMF 2.4.5 INCOMPAT Apr 1, 2003 21.02
NTSMF 2.4.7 Sep 30, 2004 22.08
NTSMF 3.1.4 Mar 15, 2009 27.01
NTSMF 4.0 Mar 15, 2011 29.03
Landmark
The Monitor for DB2 Version 2 13.06
The Monitor for DB2 Version 3.0 16.02
The Monitor for DB2 Version 3.1 20.04
The Monitor for DB2 Version 4.0 22.10
The Monitor for DB2 Version 5.0 30.02
The Monitor for CICS/ESA 1.2 - 12.12
The Monitor for CICS/ESA 1.3 - 15.01
The Monitor for CICS/ESA 2.0 - 15.06
The Monitor for CICS TCE 2.1 - 20.04
The Monitor for CICS TCE 2.2 - 20.335, 21.134 21.04
The Monitor for CICS TCE 2.3 including CICS/TS 3.1 22.08
The Monitor for CICS TCE 3.2 (almost all) 25.11
The Monitor for CICS TCE 3.2 (almost all) 27.01
The Monitor for CICS TCE 3.3 (for CICS/TS 4.1,4.2) 29.07
TMON/CICS 3.4 (for CICS/TS 5.1) 30.30-32.12
(Do not use 32.13,32.32,33.01,33.02,33.03 for 3.4)
TMON/CICS 3.4 (for CICS/TS 5.1 - Change 33.099) 33.04
TMON/CICS 4.0 (for CICS/TS 5.2 - Change 33.195) *33.09
TMON/CICS 4.1 (for CICS/TS 5.3 - Change 34.257 34.08
The Monitor for MVS/ESA 1.3 - 12.05
The Monitor for MVS/ESA 1.5 - 12.05
The Monitor for MVS/ESA 2.0 - 15.09
The Monitor for MVS/ESA 3.0 - 19.19
TMON/MVS Version 4.4 32.04
Candle
Omegamon for CICS V200 User SMF 12.05
Omegamon for CICS V300 User SMF 13.06
Omegamon for CICS V400 User SMF 16.02
Omegamon for CICS V400 type 110 segments 16.02
Omegamon for CICS V500 User SMF 18.01
Omegamon for IMS V110 (ITRF) 12.12
Omegamon for IMS V300 (ITRF) 14.04
Omegamon for IMS V550/V560 (ITRF) 25.05
Omegamon for MVS V300 13.05
Omegamon for MVS V400 13.06
Omegamon for DB2 Version 2.1/2.2 13.05
Omegamon for VTAM V160 12.04A
Omegamon for VTAM V400 15.15
Omegamon for VTAM V500 18.08
Omegamon for SMS V100/V110 12.03
CA
ACF2 6.2 was 16.04 but ABEND, ACSMFREL=0 May 2018 36.05
ASTEX 2.1 14.04
IDMS 18 32.05
IDMS 19 (INCOMPAT after PTF R084146 Change 34.164) 33.05
NETSPY 4.7 14.03
NETSPY 5.0 14.03
NETSPY 5.2 16.05
NETSPY 5.3 18.03
NETSPY 6.0 20.10 20.305
NETSPY 7.0 20.10 20.305
SAR/VIEW R11 23.07 23.196
BMC, was Boole & Babbage
APPTUNE V11R2 SMF 102 33.11 33.264
IMF 3.1 (for IMS 5.1) 12.12
IMF 3.2 (for IMS 6.1 only) 15.09
IMF 3.2 (for IMS 5.1 and 6.1+) 16.04
IMF 3.3 (for IMS 7.1 and 8.1) *22.08
IMF 4.1 (for IMS 9.1) *26.02
IMF 4.4 (for IMS 9.1) *31.08
IMF 4.5 (for IMS 11.1) (No change since 4.4) 31.08
IMF 4.6 a/k/a Mainview IMS *31.08
IMF 5.1 a/k/a Mainview IMS *34.01
IMF 5.2 a/k/a Mainview IMS 34.01
IMF 5.3 a/k/a Mainview IMS 35.03
Mainview for MQ Version 4.4 29.03
Mainview for MQ Version 5.1 30.02
Mainview for MQ Version 5.2, 5.3, 5.4 33.01
Mainview for CICS Version 6.5 (CICS/TS 5.1) 30.30
Mainview for CICS Version 6.4 (CICS/TS 4.2) 30.04
Mainview for CICS Version 6.1 26.26
Mainview Auto Operator data file 28.28
Mainview for DB2 THRDHIST file 20.20
Mainview for TCP/IP 20.20
Mainview for IP 34.??
Mainview for Batch Optimizer 19.19
Memorex/Telex
LMS 3.1 12.12A
Oracle V9, V10 24.06
SYNCSORT
2.1 33.05
1.4 33.08
Amdahl
APAF 4.1, 4.3 16.08
Velocity Software
XAMAP 3.4 22.10
XAMAP 3406 24.03
XAMAP 3.7 27.10
XAMAP 4.1 Now Renamed to ZVPS 4.1 29.07
XVPS 4.2 31.06
ZVPS 5.4 *33.07
V. Incompatibilities and Installation of MXG 37.37.
1. Incompatibilities introduced in MXG 37.37:
a- Changes in MXG architecture made between 37.37 and prior versions
that can introduce known incompatibilities.
2. Installation and re-installation procedures are described in detail
in member INSTALL (which also lists common Error/Warning messages a
new user might encounter), and sample JCL is in member JCLINSTT for
SAS Version 9.
MXG Definitions with regard to MXG Software Changes:
COMPATIBLE A change in a data record which did not alter either
COMPAT the location or the format of all of the previously-
kept MXG variables is COMPATIBLE, and you can continue
to run the old version of MXG software, which will read
the new records without error, but none of any new data
fields or any new record subtypes will be created/kept
until you install the MXG Version with this change.
INCOMPAT A change in a data record that causes the current MXG
version to fail, visibly or invisibly, with or without
error conditions or messages, and the output datasets
may contain wrong values and incomplete observations,
and/or observations may have been lost.
You MUST install the new MXG Version with this change
to process data records that have been INCOMPATIBLY
changed by their vendor.
TOLERATE In other words, the old MXG Version TOLERATES the new
data records, if they are COMPATIBLY changed.
EXPLOIT Once you use the new MXG Version to read the changed
records, all of the new fields, subtypes, etc, that are
described in this change will be created in the MXG
datasets, so the new MXG Version EXPLOITS the new data,
and you have full support of the new data records.
An MXG Version never "expires" nor "goes out of Support". When
you put in a new product/subsystem/Release/APAR that incompatibly
changed its records then you must install the current MXG Version
or at least be using the minimum level of MXG that is currently
documented in the preceding list in section IV.
COSMETIC Some Changes will start with COSMETIC. This indicates
that that change only alters a displayed value or may
be a spelling error in a label, but it is "cosmetic"
in that it ONLY affected the display, and the output
data sets created are NOT impacted by this change.
VI. Online Documentation of MXG Software.
MXG Documentation is now described in member DOCUMENT.
See also member INDEX, but it may be overwhelming.
VII. Changes Log
--------------------------Changes Log---------------------------------
You MUST read each Change description to determine if a Change will
impact your site. All changes have been made in this MXG Library.
Member CHANGES always identifies the actual version and release of
MXG Software that is contained in that library.
The CHANGES selection on our homepage at http://www.MXG.com
is always the most current information on MXG Software status,
and is frequently updated.
Important changes are also posted to the MXG-L ListServer, which is
also described by a selection on the homepage. Please subscribe.
The actual code implementation of some changes in MXG SOURCLIB may be
different than described in the change text (which might have printed
only the critical part of the correction that need be made by users).
Scan each source member named in any impacting change for any comments
at the beginning of the member for additional documentation, since the
documentation of new datasets, variables, validation status, and notes,
are often found in comments in the source members.
Alphabetical list of important changes in MXG 37.08 after MXG 36.36:
Dataset/
Member Change Description
ANAL82AU 37.214 ANAL82AU combines SERV and USER obs for TYPE82AU.
ANALCNCR 37.013 New Concurrency example counts steps and tapes.
ANALHSM1 37.038 Combined TYPE6156+HSMFSRST report, thrashing pri-mig?
ANALID 37.016 New report showing total/min/max for each SYSTEM.
ANALID 37.063 Report now shows 26.002 or 26.003 for JES2/JES3.
ANALMQ 37.039 MQ Reports replicating IBM's MQSMF program.
ANALMSUS 37.136 Powerful set of reports of SOFTWARE MSUs consumption.
ANALMSUS 37.157 MSU Consumption from TYPE89 and TYPE30 charts etc.
ANALMSUS 37.217 EXCELDEST not protected for length 0 Warning.
ANALRMF3 37.068 CF Activity Report, Structure Level, in ANALRMF3.
ASMRMFV 37.030 ASMRMFV 36.12-37.01 NOZEROCPU filter didn't.
ASMRMFV 37.178 Possible S0C4 (37.05) or S0C7 (using FDF)
ASMRMFV 37.204 Support for RMF III CRYG3 and XCFG3 and more FDF.
ASMRMFV 37.254 MXG 37.05-37.07 CPC parameter PARMS KEYWORD NULL
ASMRMFV 37.255 RMF III FDF support for PCIE and SCMG3 tables.
ASUMDB2A 37.199 Correct count of THREADS in DB2 ASUMDB2A/B/G/P/R.
ASUMMIPS 37.228 Warning message DURATM=INTERVAL conflict.
BLDSMPDB 37.106 Updated features and documentation.
BUILD005 37.195 PDB.PRINT only populated ACCOUNTn in first obs.
CICINTRV 37.210 Warning if CICINTRV INTERVAL requested can't be used.
DIFFROSC 37.256 New ROSCOE dataset ROSCOMON with all MONITORS.
DODSCRDT 37.161 New z/OS-ONLY CREATDATE variable can be created.
DODSCRDT 37.189 Spurious INVALID VALUE FOR INPUT FUNCTION
EMAIL 37.027 Example added to email SAS CondCode from ASCII SAS.
GRAFWLM 37.061 Bar charts of ZIP and ZIP eligible added.
IMACICDB 37.200 Optional CICS DBCTL var STATCTM1 too large.
JCLPDBJB 37.048 Example "BUIDPDB" creates only JOB-related datasets.
MANY 37.037 Support for SMF Manual Changes in Jan 14, 2019 Doc.
READDB2 37.042 MXG 37.01. Blank WANTONLY Cosmetic %SYSFUNC message
READDB2 37.185 APPARENT SYMBOLIC REFERENCE LDB LDB@ACT, no impact.
TECHNOTE 37.004 Reading z/OS DATA with SAS FTP Access needs RCMD
TECHNOTE 37.043 Executing MXG on ASCII, WORK needs to be local.
TECHNOTE 37.072 ODS Stat graphics procs use JAVA, memory intensive.
TECHNOTE 37.097 APAR OA65762 NEGATIVE SMF30_TIME z/OS 2.2 only.
TECHNOTE 37.110 Difference between TYPExxxx and TYPSxxxx.
TECHNOTE 37.247 Example RACF analysis, why USERID was revoked.
TYPE0203 37.216 SMF Type 2 Subtypes 1/2 GSIG/ISIG variable length.
TYPE102 37.005 Support for DB2 102 Trace IFCID 404.
TYPE102 37.051 IFCID 319 new variables created and kept.
TYPE102 37.059 Final corrections for IFCIC 319 support.
TYPE102 37.100 DB2 zPARM T106S102 variables misaligned.
TYPE102 37.240 Variable QWP4STPGS='STASTPGSAMP/N/Y/S is now INPUT.
TYPE110 37.102 All _SCICxxx Statistic SORTS deaccumulate.
TYPE110 37.102 Support for CICS/TS 5.5 new Statistics (COMPAT).
TYPE110 37.168 CICS "identity" variables not kept with UTILEXCL.
TYPE113 37.212 Support for z15 Processor SMF 113 RNI equation.
TYPE1131 37.175 New SIISPCT=STORE INTO*INSTRUCTION*STREAM*PERCENT.
TYPE119 37.003 TYP11952 SMF119ML_IP_IPV4 wrongly compressed TIRIP.
TYPE119 37.127 New variables in datasets TYP11902/TYP11994/TYP11995.
TYPE119 37.244 Subtype 3 SMF 119 not written, USERID no SMF access.
TYPE120 37.105 Support for SMF 120 WAS and LIBERTY (COMPATIBLE).
TYPE123A 37.125 Variable/format changes z/OS Connect EE 3.0 SMF 123.
TYPE123A 37.221 Support for z/OS Connect EE SMF 123 Version 2.
TYPE125 37.075 INPUT STATEMENT EXCEEDED, period missing.
TYPE1415 37.116 WPS 4.1 ONLY, U4087 ABEND,OPTIONS NOWPSSCATTERCOMP.
TYPE1415 37.172 Variable SMF14DEF='Y' if dataset is encrypted.
TYPE26J2 37.026 Local SubSystem TYPE26J2 not output, SMF6SBS NE 2.
TYPE29 37.093 Support for IMS ODBM Accounting SMF 29 Subtype 1.
TYPE30 37.058 Cosmetic: Uninitialized variable CBPERROR no impact.
TYPE42 37.034 Two more invalid SMF 42 LENSR 520 and 592 added.
TYPE42 37.135 Eight more invalid LENSR= TYPE42 subtype 5.
TYPE42 37.194 TYPE42DS variables S42DSENT/DSCMT identify zEDC.
TYPE42 37.250 SMF 42 ABEND due to SRLEN=208, now all protected.
TYPE7072 37.044 BMC CMF VERSNRMF values 792 and 794 for z/OS 2.3.
TYPE7072 37.104 Variables CECSER/CPCMODEL added to TYPE72GO.
TYPE7072 37.109 Support for z/OS 2.4 SMF Manual 04Mar19 changes.
TYPE70PR 37.046 SMF70BPS/SMF70ACS expanded for each engine type.
TYPE70PR 37.243 Vars SMF70LACM/LACA/LACB are now kept in TYPE70PR.
TYPE72GO 37.179 Variables METGOAL and PCTMETGOL were wrong.
TYPE74 37.032 TYPE749 z/EDC Divide by ZERO protection failed.
TYPE74 37.040 TYPE749 variable R7491DEFCOMPRATIO wrong value.
TYPE74 37.078 TYPE748S var R748SIID fmt $HEX4, no dupes now.
TYPE74 37.166 z/OS 2.4 Updates for TYPE7402 dataset.
TYPE79 37.242 Support for BMC SMF 70 Subtype 255 creates TYPE70FF.
TYPE80A 37.213 Support for RACF TYPE80TK TOKDANAM=AUTOLOGIN.
TYPE82 37.060 INPUT EXCEEDED SMF 82 ST 31, incorrect length.
TYPE82 37.165 TYPE8201 variables SMF82ITE/CKD/LML/USR/PKD wrong.
TYPE82 37.183 Support for SMF 82 new Audit TYPE82AU dataset.
TYPE82 37.241 Several MG082xx formats had wrong hex value.
TYPE92 37.002 TYPE9208 INPUT STATEMENT EXCEEDED, manual wrong.
TYPE92 37.017 Many non-fatal corrections were made to type 92.
TYPE92 37.085 SMF 92 Subtype 52 INPUT EXCEEDED, TRSN doc 52 bytes.
TYPE99 37.082 SMF 99 ST 12 Capacity Incr/Decr individual decodes.
TYPEAAM 37.186 Support for IBM Tivoli Advanced Allocation SMF
TYPEAXW 37.033 Support for Axway V3.3.2 2018/06/27 restructure.
TYPEBBMQ 37.073 UNEXPECTED RTIN messages, BBMQ Version 5.4 no change.
TYPEBE97 37.080A Datasets BETA9706/BETA9706D were not output to PDB.
TYPEBE97 37.197 BETA 97 Subtype 51 did not input "New Area" fields.
TYPEBETA 37.007 Support for Beta93 V6.2 Subtypes 1-3.
TYPEBETA 37.007 Support for Beta93 Version 6.2 subtypes 2 and 3.
TYPEBETA 37.114 Support for updated BETA 93 V6R2 (INCOMPATIBLE).
TYPEBETA 37.160 BETA 93 610 (back level) subtype 40/49 wrong.
TYPEBETA 37.198 BETA 93 Subtypes 12/17/30/31/55 are now output.
TYPECADK 37.219 Support for CA-DISK/Sterling DMS DSINDEX file.
TYPECIMS 37.095 New variables in TYPEDBDS (IMF from BMC).
TYPEDB2 37.035 DB2 V12 overlooked Package variables in DB2ACCTP.
TYPEDB2 37.074 QBSTBPIN always incorrectly calculated before DIF.
TYPEDB2 37.251 DB2 Storage Contraction new vars DELTAAV0/AVA/AVB.
TYPEDB2 37:014 ABEND: DB2 SMF 100 ST 1 NETEAZZA/IDAA DB2STATS.
TYPEDCOL 37.041 Support for APAR OA54879, DCDEXFLG not used FOR zEDC.
TYPEDCOL 37.069 zEDC Compression type values revised DCOLDSET/DCOLDC.
TYPEEDGR 37.015 RMM variable EDGRTIME had missing values.
TYPEFOCU 37.215 Support for FOCUS Version 7.7 USER SMF Record.
TYPEHSM 37.076 Support for HSM FSR Record Unix filename added.
TYPEIMS 37.103 Support for IMS Log Records 5607/5610/5904/5950.
TYPEIMS 37.176 Support for IMS LOG TYPE '02'x.
TYPEMAR 37.181 Support for Hitachi MAR Mainframe Analytics 9.1
TYPEMDM 37.015 RMM variable EDGRTIME had missing values.
TYPENDM 37.047 NDM-CDI dataset NDMCT var NDMCPU 256 times too large.
TYPENDM 37.113 NDM-CDI 24 byte short record INPUT EXCEEDED.
TYPERMFV 37.001 Some RMF III ZRBASI fields blank/wrong in 36.36.
TYPERMFV 37.055 CFACT Coupling Facility Structure Activity Report.
TYPERMFV 37.067 Support for RMF III PCI/SCM/ZFX segments 4 datasets.
TYPERMFV 37.080 SVPCNM and RPRTCLAS added to all RCD datasets.
TYPERMFV 37.167 z/OS 2.4 Updates for RMF MONITOR III datasets
TYPERMFV 37.246 Variables PCIFTET/PCIFTQT in ZRBPCI incorrect.
TYPESAPZ 37.222 Support for SAP Z Connector USER SMF record.
TYPESTC 37.018 Support for STC HSC Subtype 32 and 33 new datasets.
TYPETLMS 37.192 Support for TLMS creates two new datasets.
TYPETPX 37.107 Misaligned TPXETIME reported as 8a Oct 27, 1935.
TYPEVM 37.130 New VM Account datasets supported.
TYPEVMXA 37.012 ABEND: ZVM MONWRITE NEW 40061802 Service Level.
TYPEVMXA 37.028 Support for z/VM 7.1 (INCOMPAT, BROKEN CONTROL).
TYPEVMXA 37.084 z/VM VXPRCAPM dataset vars CMB10C0-X4 wrong values.
TYPEWSF 37.083 Logic revised to use OBJT/ACT for Input choice.
TYPEWSF 37.111 Final revisions for WSF/EOS WSFAUDIT AUDACT/OBJN.
TYPEXAM 37.081 Analyzing VPS USER dataset, must use INTORSUM='SU'.
UTILMISS 37.053 Utility to remove all variables that are all missing.
VMACEDGR 37.225 RMM/EDGR dataset EDGRXEXT RDPHYSIZE too small.
VMACSMF 37.249 Using FTP ACCESS from z/OS fails with JFCB Issue.
VMXGALOC 37.021 New parms YR2KEEP and BASEYEAR for Yearly PDB.
VMXGALOC 37.224 Note if you added VMXGALOC to your IMACINIT.
VMXGDSN 37.191 RMM/EDGR in VMXDSN had zero obs for TAPES/TAPEDSNS.
VMXGOPTR 37.252 Restores ORIGINAL, used to always reset if changed.
VMXGPRAL 37.006 Unbalanced parens in variable label, non fatal.
VMXGSUM 37.245 KEEPALL=NO with INCODE= could cause UNINIT message.
VMXGUOW 37.011 Enhanced for each CICSTRAN-only PDB.ASUMUOW.
CONFIMXG 37.267 DO NOT USE ENCODING=EBCDIC1047, use OPEN_ED-1047.
ASUMMSUS 37.265 ERROR PARM OPTNE NOT DEFINED, typo.
TYPE99 37.264 SMF 99 Subtype 2 TYPE99_2 new variables added.
TYPE42 37.261 SMF 42 Subtype 1 TYPE42DS INPUT STATEMENT EXCEEDED.
TYPEIDMS 37.260 IDMS dataset IDMSTAS USER Fields were misaligned.
TYPEIMS 37.259 IMS SYNCPOINT Log Records IMS5937/5938 keeps all vars
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 37.272 WERE IN MXG 37.37 DATED Jan 6, 2020 ========
Change 37.272 If you did not specify EXCELDEST you could get WARNINGs:
ANALMSUS Argument 2 to macro function %SUBSTR is out of range
Jan 5, 2020 Argument 3 to macro function %SUBSTR is out of range
Change 37.271 %BUILDIMS can create all current MXG IMSLOG datasets and
BUILDIMS REPORT=YES (default) prints counts of bytes/records for
FORMATS each LSUBCODE; REPORT=ONLY will produce only the report.
VMACCIMS Not all IMS Log Records create MXG datasets; many records
VMACIMS are for database recovery with no performance metrics,
VMXGINIT but support will be added upon request and test data.
Jan 5, 2020 // EXEC MXGSASV9
//IMSLOG DD DSN=YOUR.IMS.LOG.DATA,DISP=SHR
//PDB DD DSN=YOUR.IMS.PDB,DISP=(,CATLG),SPACE=....
//SYSIN DD *
%BUILDIMS(REPORT=YES,
WANT= F9 FA 56FA,
IMSVER=14.1,
START= '01JAN2020:08:00:00'DT,
END = '01JAN2020:10:00:00'DT);
Note: REPORT=YES reports all IMS Log Records in input and
is not impacted with WANT/START/END/IMSSYSTEM selections.
See comments in member BUILDIMS for documentation.
Thanks to Randy Hewitt, DXC. USA.
Change 37.270 A TECHNOTE documenting how to recover from MXG ABENDS.
ADOCABND
Jan 1, 2020
Change 37.269 TYPE71 variables SMF71PIS/SMF71POS, 4K TO/FROM SCM are
VMAC71 accumulated fields, so they are correct only when TYPS71
Dec 23, 2019 (or BUILDPDB) is used, since the deaccumulation is done
in the _STY71 Data Set Sort Macro, and will always be
missing values in the first obs from each SYSTEM.
Thanks to Graham Harris, RBS, ENGLAND.
====== CHANGES THRU 37.268 WERE IN MXG 37.09 DATED Dec 20, 2019 ========
Change 37.268 Variable SM116EVT in MQMACCTQ and MQMQUEUE datasets was
VMAC116 not set for Task Terminated; test for '21'x was wrong and
Dec 20,2019 changed to four byte test for '200800000'x for Terminate
and '00080000'x for Initiate.
Thanks to Scott Barry, SBBWORKS, INC, USA.
====== CHANGES THRU 37.267 IN EARLY ADOPTER MXG 37.09 DATED Dec 20, 2019
Change 37.267 z/OS: ENCODING=EBCDIC1047 corrupts data; OPEN_ED-1047 is
CONFIMXG and has been REQUIRED for MXG, and is the z/OS Default.
VMXGINIT With ENCODING=EBCDIC1047 character variables with '15'x
Dec 12, 2019 byte value are changed to '25'x, and byte values of '25'x
are changed to '15'x! Those are the New Line & Line Feed
characters that swap. You can display your options with:
PROC OPTIONS OPTION=ENCODING;
PROC OPTIONS OPTION=LOCALE;
PROC OPTIONS OPTION=NLSCOMPATMODE;
-You can add an ENCODING=OPEN_ED-1047 statement in your
CONFIGV9/CONFIMXG/CONFIGxx tailoring member, or you can
set with // EXEC SAS94,OPTIONS='ENCODING=OPEN_ED-1047'
-Not only can character bit tests and formatted values be
wrong, some numeric TODSTAMP datetime fields are created
from character variables when shorter than 8-bytes, using
DATETIME=INPUT(CHAR6!!'0000'x),TODSTAMP8.);
Changing the second byte from '15'x to '25'x is 30 days.
-VMXGINIT is enhanced to print error message if ENCODING
option is not OPEN_ED-1047.
-And, OPEN_ED-1047 does work with NONLSCOMPATMODE, And SAS
in 2016 without documenting it, said they were NOT going
to remove NLSCOMPATMODE nor NONLSCOMPATMODE options.
-If you depend on NLS characters, the CONFIMXG option was
created by SAS to support NLS sites, see that member.
Thanks to Andrew Gadsby, SAS UK, ENGLAND.
Thanks to Andy Knight, SAS UK, ENGLAND.
Thanks to Stephen Hoar, Lloyds Banking, ENGLAND.
Thanks to Mark Tomlinson, Lloyds, ENGLAND
Change 37.266 Parameters DATETIME= and INTERVAL= were not specified in
TRNDDB2P VMXGSUM invocation, causing little or no summarization.
TRNDDB2R
Dec 11, 2019
Thanks to Marybeth Delphia, CPA.TEXAS.GOV, USA.
Change 37.265 If you did not specify PDBOUT the correct dataset(s)
ANALMSUS names were not generated.
Dec 11, 2019 -The use of substitution macros in IMACKEEP made ANALMSUS
fail if you tried to run it twice in a single SAS
session. Those affected macros are now nulled out.
-OPTIONS NOBYLINE was set but there were no BY values
in the TITLE statements - there are now.
-MXG 37.09 ONLY. ERROR KEYWORD PARM OPTNME NOT DEFINED,
now correctly spelled as OPTNAME.
-Eliminated a WARNING about SUBSTR out of range when
EXCELDEST was NULL.
Thanks to Jan Tielemans, KBC, BELGIUM
Change 37.264 SMF 99 Subtype 2 TYPE99_2 dataset did not input the many
VMAC99 variables that had been added in z/OS 2.3.
Dec 15, 2019
Thanks to Betty Wong, Bank of America, USA.
Change 37.263 READDB2 didn't create the T102SSSS and T102S000 datasets
READDB2 when a numeric IFCID was in the IFCIDS= parameter. Now,
Dec 10, 2019 it does unless T102SSS/T102000 are set to NO.
Logic to parse strings for each dataset was simplified
and relocated to a separate macro to simplify ongoing
maintenance.
Change 37.262 Replaced by Change 37.271.
BUILDIMS
Jan 1, 2020
Change 37.261 TYPE42 INPUT STATEMENT EXCEEDED subtype 6 due to a short
VMAC42 S42DSSNL=72 segment when SMF manual had 80, probably due
Dec 7, 2019 one of the APARs in Change 37.250.
-Subtype 5, LENSR=208 records were deleted (Change 37.260)
with log message, but now they are output and the extra
48 bytes are skipped.
Thanks to Richard Haynes, Blue Cross Blue Shield of Kansas, USA.
Change 37.260 IDMS dataset IDMSTAS fields were misaligned because the
VMACIDMS new variable TASBJBID was inserted in the record.
Dec 6, 2019
Thanks to Paola Rosero, Government of Quebec, CANADA
Change 37.259 IMS SYNCPOINT Log Records IMS5937 and IMS5938 only kept a
VMACIMS small number of variables, now all fields in the record
Dec 16, 2019 are kept.
Thanks to Randy Hewitt, DXC. USA.
Change 37.258 Earlier releases of the %SCAN function of the %MACRO
READDB2 COMPILER returned the input string if there was only
Dec 1, 2019 a single string for the delimiter that was used, which
required extra MXG logic to correct, and which could fail
if AAAA was used, with no BBBB, and there was a CCCC but
no space between the two slashes, causing CCCC to be seen
as the BBBB argument, raising a WORD2 macro error. Now,
SAS returns the string correctly so the excess MXG logic
could be removed along with this (rare) exposure.
-If all you requested with WANTONLY was DB2ACCT DB2ACCTW
was also produced.
Thanks to Ervin Claxon, CSX, USA.
Change 37.257 This member will work but is obsolete and has been
TRNDRMFN replaced by TRNDRMFI. The biggest difference is in the
Nov 29, 2019 input dataset. This member uses PDB.RMFINTRV which would
only get a single days PDB as input while the TRNDRMFI
member uses WEEK.RMFINTRV which would be all of the days
of the week.
Thanks to Wayne Bell, UNIGROUP, USA.
====== CHANGES THRU 37.256 WERE IN MXG 37.08 DATED Nov 26, 2019 ========
Change 37.256 ROSCOE new dataset PDB.ROSCOMON contains an observation
DIFFROSC for each of these MONITOR values:
Nov 19, 2019 ALL AMS ATT CON CPL DIS DSF ETS HEX IMP
JCL LIB LOO OTH OUT PUR SOR UTI VTO ZAP
and variable MONITOR contains the monitor acronym.
The Monitors AUD AWS RPS SHU STA VPE are not included,
as they have always had a separate ROSCOxxx dataset.
Thanks to Linda S. Berkley, DISA, USA.
Change 37.255 -FDF support added for the RMF Monitor III PCIG3 (PCIE
ADOCRMFV Activity Data Table) and SCMG3 (Storage Class Memory Data
ASMRMFV Table). General ASMRMFV support for these tables already
Nov 19, 2019 existed.
-Improved ASMRMFV FDF handling of 16 byte binary
Fields using Extended Floating Point arithmetic.
-Minor corrections made to ASI, CPD, GEI, OPD, SPG, and
SSHG3 FDF Variable Name Tables.
-Main storage reduction made for FDF Alias and True
Name entries in Variable Name Tables.
-FDF now uses only 1 pattern table for all character
Fieldnames instead of a pattern table for every character
Fieldname when a user character pattern compare occurs.
-Message RMFV088I now shows XFP (Extended Floating Point)
tag for 16 byte quadword fields during FDF processing.
-Floating point fields were not always scaled by FDF
when required prior to comparisons to a user value.
-When the maximum 32K LRECL output length to RMFBSAM would
be exceeded in PROCCPU, PROCCFI, and PROCRCD subroutines
the entire RMF III table will now be skipped. This should
be a very rare event.
-Section 12 Messages updated in ADOCRMFV.
-These documentation sections in member ADOCRMFV are
added for new FDF support:
40 Filtering The PCIG3 Table
41 Filtering The SCMG3 Table
Remaining existing section numbers are incremented by 2.
Change 37.254 -MXG 37.05-37.07. When CPC parameter was specified,
ASMRMFV Message RMFV057I PARMS KEYWORD VALUE IS NULL printed
Nov 16, 2019 and neither CPCDB nor CPUG3 tables were selected.
-PCIG3 table support was not checking the z/OS release.
Only z/OS 2.3 and higher releases are supported.
Inputting a PCI table from a lower release results in an
MXG build failure. PCIG3 tables from older releases will
now be ignored. Applies to MXG Releases 37.03 and up.
Change 37.253 Cosmetic. Labels were blank for these variables/datasets
ASUM70PR STCVSM32 variable SMLSSMF32
VMAC102 ASUMCEC variable CPUICFTM
VMAC113 TYPE1131 EXTND22x variables had COUNTER 225 UNDEFINED
VMAC120 FOCUSMSO variables FOCUMEMA,FOCUMEMB
VMAC26J2 TYPE26J2 variable SUBS
VMAC82 TYPE82 variable SMF82_TAK_KEY_EVENT
VMACSTC TYPE120 variable SMF1209HW
Nov 18, 2019 T102S106 variable QWO4ADM2-QWO4SADM
Thanks to Chris Weston, SAS ITRM Development, USA.
Change 37.252 During QA testing we found that 250 members changed a
VMXGOPTR SAS system option but did not return that option to its
MANY original value, which could impact subsequent programs
Nov 18, 2019 in that SAS job step or SAS Session. Previously, with
NEWVALUE=ORIGINAL, VMXGOPTR reset the option to its
value PRIOR to the LAST VMXGOPTR invocation, but now
it is restored to the original value of the SAS Session
or batch STEP, and all 250 members now have pairs of
VMXGOPTR to change and to then reset options to their
ORIGINAL value.
Example of use of VMXGOPTR with this change with the
original option DSNFERR having value DSNFERR:
%VMXGOPTR(OPTNAME=DSNFERR,NEWVALUE=NODSNFERR);
this code executes with NODSNFERR
%VMXGOPTR(OPTNAME=DSNFERR,NEWVALUE=NODSNFERR);
this code executes with NODSNFERR
%VMXGOPTR(OPTNAME=DSNFERR,NEWVALUE=ORIGINAL);
subsequent code executes with DSNFERR.
With the original VMXGOPTR, the subsequent code was
executed with NODSNFERR, which was wrong.
-The report members also set TITLE and FOOTNOTEs to null
to prevent carry-forward for following program reports.
Change 37.251 New DB2STATS variables DELTAAV0/AVA/ACB are negative when
VMACDB2 storage contraction starts, and the value is the number
Nov 7, 2019 of pages needed. Values for Available, DBM1 and DIST are
created. IBM reports, with this ZPARM (xxxxxxxx) "AUTO",
contraction will begin when paging is detected and DB2
will try to bring a system to a point where paging is
minimal or non-existent. DB2 will enter a full
contraction mode if 100% of REALSTORAGE_MAX is hit, when
message DSNS0003I is issued, and the discard will start,
if PM99575 is applied. DB2STATS variable QSST_RSMAX_WARN
counts the number of times the REALSTORAGE_MAX Warning
was reached in each interval.
Thanks to Sieghart Seith, FICUCIA, GERMANY.
Change 37.250 SMF 42 Subtype 5 APAR OA52133 causes ABEND with LENSR=208
VMAC42 in SMF42SRL and a 208 byte segment, APAR OA54663 corrects
Nov 7, 2019 setting SMF42SRL=160 with only a 160 byte segment. New
length was introduced by the usermod ++APAR for OA52133,
when it was still in development, and before OA54663.
-Below Change Text was insufficient to protect for 208,
and logic was completely revised in Change 37.261, 37.09.
To protect for future invalid lengths to avoid ABENDs,
MXG logic first tests for a delta of 160 bytes between
Storage Class Name fields and uses LENSR=160 if true,
or the SMF42SRL value is tested for a valid delta, and
uses that value if true. Otherwise, the record is
skipped with an MXGNOTE on the log.
Thanks to Hiroshige Koshigoe, FTB CA, USA.
Thanks to Sarel Swanepoel, SARS, SOUTH AFRICA.
Change 37.249A The FTP ACCESS method executing on z/OS to read z/OS data
CHCK4FTP fails because the expected JFCB/DSCB tokens do not exist
Nov 22, 2019 with the SAS FTP Access method.
-Most MXG use of the FTP ACCESS method executes on ASCII
to read z/OS SMF data, which does not have this exposure.
-There are two ways to use z/OS to z/OS ftp:
Using SMF as an example DDNAME/INFILE name:
FILENAME SMF FTP "'MYSMFDATA'" HOST='IP ADDRESS'
USER='USERNAME' PASS='PASSWORD' S370VBS LRECL=32760
RCMD='SITE RDW';
%LET MXGJFCB=;
%LET MXGDSCB=;
%INCLUDE SOURCLIB(BUILDPDB);
or
FILENAME SMF FTP "'MYSMFDATA'" HOST='IP ADDRESS'
USER='USERNAME' PASS='PASSWORD' S370VBS LRECL=32760
RCMD='SITE RDW';
%CHCK4FTP(DDNAME=SMF);
%INCLUDE SOURCLIB(BUILDPDB);
-The %CHCK4FTP(DDNAME=yourdd); in example two validates
that that DDNAME is allocated and then detects the engine
type, and for FTP, MXGJFCB and MXGDSCB macro variables
are blanked to prevent the JFCB/DSCB error.
Thanks to Mike Martin, NC State Employee's Credit Union, USA.
Change 37.249 %DSCRDT now defaults to a null value and is enabled by
DODSCRDT using %INCLUDE SOURCLIB(DODSCRDT); in your //SYSIN.
VMXGINIT -These INFILE names can be enabled for these TYPExxxx:
Nov 26, 2019 TYPExxxx: TYPECTLT TYPEDCOL TYPEEDGR TYPERMFV
INFILE: CONTROLT DCOLLECT EDGHSKP RMFBSAM
TYPExxxx TYPETMS5 TYPEIMS
INFILE: TMC IMSLOG
See change 37.161 for the original addition.
Change 37.248 Unused Change Number.
Nov 4, 2019
Change 37.247 Example analysis to track down why USERID was revoked.
TECHNOTE
Nov 4, 2019
SUMMARY: Userid for an inbound FTP process had the wrong password,
eventually userid was revoked, but the MVS RACF SYSLOG message was
misleading, as it was expected to be a VTAM LU, but it is the value
"A86CF325" in message that matches the RACFTERM:
MVS SYSLOG excerpt
ICH408I USER(FCADOU4 ) GROUP(ADOFTP ) NAME(FTPS ID CDS T4) 546
LOGON/JOB INITIATION - INVALID PASSWORD ENTERED AT TERMINAL A86CF325
IRR013I VERIFICATION FAILED. INVALID PASSWORD GIVEN.
MXG TYPE8001
RACF*USER*NAME [RACFUSER] FCADOU4
EVENT*CODE*QUALIFIER [RACFQUAL] 101:INVALID PASSWORD
RACF/VTAM*TERMINAL*NAME USED [RACFTERM] A86CF325
MXG VMAC80A: INPUT RACFTERM $EBCDIC8. /*@43 SMF80TRM*/
RACF MACROS and INTERFACES Documentation
SMF80TRM 8 EBCDIC Terminal ID of foreground user (zero if not avail)
MXG TYP11902
CONNECTION*ESTABLISHMENT [TTSTIME] 26OCT2019:09:03:17.99
TIMEWAIT OR*LASTACK [TTETIME] 26OCT2019:09:03:18.22
REMOTE*IPV6*ADDRESS [TTRIPV6] 0000:0000:0000:0000:0000:FFFF:A86C:F325
(matches RACFTERM from Type 80, and the MVS RACF ICH408I message)
Change 37.246 Dataset ZRBPCI variables PCIFTET and PCIFTQT were input
VMACRMFV incorrectly as milliseconds (PIB8.3) units, but they are
Oct 31, 2019 both microseconds (PIB8.6), with TIME13.3 print format.
Change 37.245 If you KEEPALL=NO or %LET MXGKEEP=NO and your VMXGSUM
VMXGSUM invocation had an INCODE= you might have gotten an
Oct 30, 2019 UNINIT message on one or more variables in the INCODE
that were not also referenced in one of the parameters.
This change modifies KEEPALL to YES if INCODE is used.
Change 37.244 Observations from Subtype 3 SMF 119 records were not
VMAC119 output because some of the Production UserIds did not
Oct 30, 2019 have read access to the SMF unix Facility, which is
required for SMF record to be written.
Thanks to Aslyona Bertneski, Express-Scripts, USA.
Change 37.243 Variables SMF70LACM SMF70LACA and SMF70LACB are now kept
VMAC7072 in TYPE70PR dataset to aid in SCRT comparisons. They
Oct 28, 2019 have been in TYPE70 since Version 33.
Thanks to Ken Deering, Compuware, USA.
Change 37.242 Support for BMC SMF79 Subtype 255 creates new dataset:
EXTY79FF DDDDDD Dataset Description
FORMATS TY79FF TYPE79FF BMC SMF 79 SUBTYPE 255
IMAC79
VMAC79
VMXGINIT
Oct 28, 2019
Thanks to Randy Hewitt, DXC, USA.
Change 37.241 Several MG082xx format for SMF 82 records had wrong hex
FORMATS value displayed, but the text description was correct.
Oct 27, 2019
Thanks to Ron Rust, Worldpay, USA.
Change 37.240 DB2 Variable QWP4STPGS='STATPGSAMP/N/Y/S' was not INPUT
VMAC102 in T102S106 dataset.
Oct 24, 2019
Thanks to Lai Fai Wong, Bank of America, USA.
====== CHANGES THRU 37.239 WERE IN MXG 37.07 DATED Oct 22, 2019 ========
Change 37.239 Support for z/15 SMF113 records new CRYPTO17-CRYPTO20
VMAC113 counters (ERROR: ARRAY SUBSCRIPT OUT OF RANGE ABEND),
Oct 22, 2019 only if you had enabled the HIS CRYPTO counters.
Thanks to Steven W Erkkila, USBANK, USA.
Change 37.238 Support for NDM-CDI 'S#' record type is now output in
VMACNDM existing NDMAE dataset.
Oct 16, 2019
Thanks to Rob D'Andrea, Royal Bank of Scotland, ENGLAND.
Change 37.237 Cosmetic. STARTIME syntax modified to eliminate notes
ASUM42DS from VMXGSUM internal invocation.
Oct 15, 2019
====== CHANGES THRU 37.236 WERE IN MXG 37.07 DATED Oct 14, 2019 ========
Change 37.236 Support for z/15 SMF 113 records new EXTEND256-EXTEND287
ANALSIIS counters, with corresponding ARRAY size increases, CAUSED
VMAC113 ERROR:ARRAY SUBSCRIPT OUT OF RANGE ABEND. In ANALSIIS,
Oct 14, 2019 SORT order was corrected.
Thanks to Tony Curry, BMC, USA.
====== CHANGES THRU 37.235 WERE IN MXG 37.07 DATED Oct 12, 2019 ========
Change 37.235 CRITICAL ERROR if BLDSMPDB+VMXGALOC is used on ASCII.
BLDSMPDB An ending comma in line 949 of BLDSMPDB caused
Oct 12, 2019 ERROR: All positional parameters must precede keyword.
Remove the comma in line 949 of BLDSMPDB.
====== CHANGES THRU 37.234 WERE IN MXG 37.07 DATED Oct 9, 2019 ========
Change 37.234 DB2 variables added to DB2ACCT and DB2STATS datasets:
VMACDB2 QXSTEHLST='EXECUTION*HISTORY.LOST'
Oct 8, 2019 QXSTHVLST='HV*RECORDING*HISTORY*LOST'
QXRFMIAP='RID*LIST*PROCESSING*NOT USED'
Thanks to Warren Cravey, Fidelity, USA.
Change 37.233 CICS variable ABCODE created by UTILEXCL was the concat
UTILEXCL of ABCODE113!!ABCODE114, but using INPUT ABCODE $EBCDIC8.
Oct 8, 2019 eliminated the need for that statement, which caused
ERROR: The name ABCOD113§§ABCOD114 is not a valid SAS name
when UTILEXCL was run on Linux. A track with SAS will be
opened to resolve that error, because double exclamation
points are used 6900 times in MXG for concatenation, but
this code matches the VMAC110 code and should have been
there some time ago, just for consistency, but now for
circumvention.
Change 37.232 -If CICS Dynamic Transaction Routing, DTR is used, the
VMAC110 CICSTRAN has PROGRAM='########' which historically was
VMXGINIT ONLY used when there was an invalid transaction name
Oct 9, 2019 typed in by the user, which caused many fields to be
missing values, so MXG instead output the transaction
to the CICSBAD dataset. But now, if you enable DTR,
the observation count in CICSTRAN may drop dramatically
by the number of obs output in CICSBAD. There does not
appear to be a flag that identifies DTR was used.
You will need to set
%LET MXGCICSDTR=YES;
to cause MXG to output ALL transactions with '########'
values to CICSTRAN instead of to CICSBAD.
-The original test for CICSBAD also tested TRANFLAG for
a bit that indicated invalid times in the record, but
IBM removed that bit, so that bit test was removed.
-Changes 23.312 and 24.155 document PROGRAM='########';
Thanks to Scott Barry, SBBWORKS, INC, USA.
Change 37.231 z/VM 6.4.18.2 BROKEN CONTROL RECORD in IODVSW extended
VMACVMXA INPUT for SKIP GE 20 was missing the @ at the end.
Oct 7, 2019
Thanks to Rob D'Andrea, Royal Bank of Scotland, ENGLAND.
Change 37.230 If you specified DURATM=INTERVAL with INTERVAL= values of
VMXGSUM TWOMIN THREEMIN or TWENTYMIN the DURATM was not created.
Oct 6, 2019
Change 37.229 Doc only change - corrected spelling and syntax.
ANALCAPD
Oct 5, 2019
Change 37.228 -Format $MGRMIPS is updated for the z/15 processor type
ASUMMIPS 8561 to map the CPCFNAME ('8561-401') to the MIPS per MSU
FORMATS value (267), used in ASUMMIPS and other 'MIPS' calcs.
Oct 5, 2019 -Protection for CPCFNAME not found in $MGRMIPS format now
gracefully terminates with a message from ASUMMIPS.
Change 37.227 QLACRLNU was not added to DB2ACCTB nor DB2ACCTG causing
VMACDB2 UNINIT messages in ASUMDB2B and ASUMDB2G.
Oct 4, 2019
Change 37.226 TYPE 120 dataset TY120100 variables SM120RULEXCALLS (was
VMAC120 always -1) and SM120RULEXFSUM were in the wrong order in
Oct 3, 2019 the INPUT statement.
Thanks to Scott Barry, SBBWorks, Inc., USA.
Change 37.225 RMM/EDGR dataset EDGRXEXT variable RDPHYSIZE was not
VMACEDGR multiplied by the RDFACTOR(KB/MB/GB) so it was too small.
Oct 1, 2019 The value was correct in the other EDGR datasets.
Thanks to Bradley Leis, TELUS, CANADA.
Change 37.224 If you correctly added VMXGALOC to your IMACINIT so that
VMXGALOC you would always have the current PDBs allocated to each
BLDSMPDB SAS/MXG session and changed the number of PDBs (daily,
Oct 1, 2019 weekly, monthly, etc.) in BLDSMPDB you could possibly
wipe out old directories unintentionally. VMXGALOC now
detects that it has already run and unless the new
INVOKEBY= parameter is set to bldsmpdb sets the READONLY
parameter to YES to prevent deletion of old directories.
VMXGALOC in IMACINIT should always specify READONLY=YES
so this is only insurance that will produce and MXGWARN
message and set condition code 8.
Change 37.223 DB2 IFCID 376 variable QW0376SINR (statement count) was
VMAC102 not kept in 8 bytes, causing small differences in the
Sep 30, 2019 MXG value (222986880) vs IBM (222986962) = 82 smaller.
With stored length of 8 bytes, the values match.
Thanks to Warren Cravy, FMR, USA.
Change 37.222 Support for SAP Z Connector USER SMF Record creates
EXSAPZCO DDDDDD DATASET DESCRIPTION
IMACSAPZ SAPZCO SAPXCO SAP Z CONNECTOR
TYPESAPZ
TYPSSAPZ
VMACSAPZ
VMXGINIT
Sep 30, 2019
Thanks to Nestor D Rossi, BancoGalica, ARGENTINA.
Change 37.221 Reserved Change.
Change 37.220 If you specified SYNC59=NO (the default) the intervals
VMXGDUR always landed (for example with INTERVAL=QTRHOUR) on the
Sep 26, 2019 quarter at 0 15 30 45 minutes. If you wanted to keep the
actual intervals you would have needed to subtract 1
minute. VMXGDUR will now do that for you with the new
USE59 parameter. YES will subtract 60 seconds from the
calculated value and get the QTRHOUR back to 59 14 29 44
minutes. You can implement this globally be inserting in
IMACINIT: %LET SYNCTO59=YES;
The purpose of SYNC59 was always to put the start of
intervals on the hour since that usually makes more sense
on management reports and graphs. But if you choose to
use this on interval data without modifying the FLORCEIL
parameter to CEIL and you use SYNC59=NO you will be
getting the STARTIME of the prior interval rather than
the current interval since 12:59 with the FLORCEIL=FLOOR
set resolves to 12:45:
FLOOR('12:59't/900)*900;
Examples using default of SYNC59=NO
FLORCEIL=FLOOR
TIMESTMP=29AUG18:12:59:00 datetime=29AUG2018:12:45:00
FLORCEIL=FLOOR USE59=YES
TIMESTMP=29AUG18:12:59:00 datetime=29AUG2018:12:44:00
FLORCEIL=CEIL
TIMESTMP=29AUG18:12:59:00 datetime=29AUG2018:13:00:00
FLORCEIL=CEIL USE59=YES
TIMESTMP=29AUG18:12:59:00 datetime=29AUG2018:12:59:00
Thanks to Randy Hewitt, DXC, USA.
Change 37.219 Support for CA-DISK/Sterling DMS file DSNINDEX.
EXCAARCH INFILE is CADISK, creates two new dataset:
EXCADISK DDDDDD DATASET DESCRIPTION
FORMATS CADISK CADISK DSNINDEX ARCHIVED OR BACKED UP.
IMACCADK CAARCH CAARCH ARCHLOGS ARCHIVED DATASETS.
TYPECADK -Invalid LEAP YEAR dates (year 2025, day 366) can't be
TYPSCADK set to a numeric value, variable DSNDS1EDLEAP will now
VMACCADK have the invalid date as character (2026366).
VMXGINIT
OCT 2, 2019
OCT 30, 2019
Thanks to Pierre-Pascal Joulin, SOCGEN, FRANCE.
Change 37.218 DURATM=INTERVAL conflicted with DURATM in the SUM=
ASUMMIPS parameter which raised an MXGWARN message. DURATM
SEP 26, 2019 was removed from the SUM= list.
MXGWARN: DURATM=INTERVAL WAS SPECIFIED BUT THE DURATM
MXGWARN: VARIABLE CONTAINS NON-MISSING VALUES. THIS
MXGWARN: MAY BE AN ERROR CONDITION. VALUE OF DURATM
MXGWARN: IN THE OUTPUT DATA IS: 0:15:00.00
Change 37.217 A test of EXCELDEST was not protected for length of 0
ANALMSUS and generated a WARNING message but still ran albeit
Sep 26, 2019 with condition code 4.
Thanks to Robert Sample, TOMY, USA.
Change 37.216 SMF Type 2 Subtypes 1 and 2 fields SMF2GSIG and SMF1ISIG
TYPE0203 were documented as 512 bytes long, but IBM updated their
Sep 26, 2019 doc that they are variable length fields, so the INPUT
is revised to use the SMF2GSIGLEN and SMF2ISIGLEN fields
to input the variable length data.
Change 37.215 Updates for FOCUS USER SMF record changes version 7.7.
ADOCFOCU -VMACFOCU was changed to accommodate a change by IBI in
FORMATS their SMF records somewhere around version 7.7 of
VMACFOCU WebFOCUS (vendor support wasn't clear about when the
Sep 25, 2019 change happened).
Nov 14, 2019 -User ID fields were changed from 8 to 20 bytes and a
'security provider' value was prepended to each user ID
with a '\' as separator. These changes caused other
fields in the SMF record to be in different columns than
they were previously. Fields were also added for zIIP
time and zIIP-on-CP time.
-There is no version field in the SMF record, so this code
uses the record subtype and the length of the record to
determine which record format to use. There are four
subtypes - 1,2,4, and 5, which are Logon, Logoff, Begin
Query, and End Query, respectively. Old format records
are shorter than new format records, so the code checks
to see if a record is shorter than the new format, and if
it is, uses the old format. Older format records will
have missing or blank values for old variables that no
longer exist in the new records.
-Each user ID variable (Logon User ID, Security User ID,
Pooled User ID) now has a corresponding security provider
variable. The provider will be blank if an older format
record was used.
-zIIP time and zIIP-on-CP time variables are normalized,
according to the vendor. We are not sure of the accuracy
of these fields, and would not count on them for serious
reporting.
-Members ADOCFOCU and VMACFOCU were changed to reflect
these changes. Some labels were changed to reflect that
these records are produced by the vendor's WebFOCUS
product in addition to the FOCUS MSO (Multi-System
Option). In addition two new values are added in the
new MGFOCTY format to reflect subtypes 4 (Begin Query)
and 5 (End Query).
Thanks to Tim Hare, Hare Systems, USA.
Change 37.214 ANAL82AU combines the separate SERV and USER observations
ANAL82AU into one observation with two sets of variable names.
Sep 25, 2019
Thanks to Alexander Bitter, Worldpay, USA.
Thanks to Rib Rust, Worldpay, USA.
Thanks to Brian Bowling, Worldpay, USA.
Change 37.213 Support for RACF 80A TOKDANAM=AUTOLOGIN creates new
VMAC80A variable TOKAUTOLOGIN in TYPE80TK dataset.
Sep 24, 2019
Thanks to Andrew Krink, Northern Territory Government, AUSTRALIA.
Change 37.212 Support for z15 processor SMF 113 RNI equation changed to
ASUM113 RNI=2.9*(0.45*L3P+1.5*L4LP+3.2*L4RP+6.5*MEMP)/100;
VMAC113
Sep 23, 2019
Change 37.211 Updates to TYPERMFV.
IMACRMFV -Variable PCIFTYPHEX added to ZRBPCI dataset.
VMACRMFV -Support for XCFG3 segment creates three datasets
Oct 2, 2019 that contain the same data as RMF I XCF records:
DDDDDD DATASET DESCRIPTION SAME DATA
ZRBXCG ZRBXCG RMF III XCF GROUP TYPE74ME
ZRBXCP ZRBXCP RMF III XCF PATH TYPE74PA
ZRBXCS ZRBXCS RMF III XCF SYSTEM TYPE74SY
Change 37.210 If the default requested INTERVAL=_CICINTRV (HALFHOUR)
CICINTRV doesn't match your actual CICS Statistics interval, or
Sep 20, 2019 isn't an exact integer multiple, MXG printed a warning
that the resultant CICINTRV dataset was invalid, as it
built using your _CICINTRV value.
Now, your _CICINTRV value is compared with the maximum
DURATM in your data, and if your requested _CICINTRV can
be honored it will be; otherwise the interval of your
data is used, and a warning of the change is printed.
Change 37.209 Spurious log messages INVALID VALUE FOR INPUT FUNCTION
DODSCRDT that occur only if INFILE is on TAPE are harmless with no
Sep 19, 2019 impact on the output MXG datasets. Thought removed by the
Aug 30 MXG 37.06 Change 37.189, they weren't until this.
Thanks to Betty Wong, Bank of America, USA.
Change 37.208 ID & 89 added to the list of product suffix found in your
UTILBLDP USERADD= argument that are already in BUILDPDB=YES and
Sep 19, 2019 would have caused an error if honored.
Thanks to Randy Hewitt, DXC, USA.
Change 37.207 -New format MG110XL is created for UCICSCNT report so you
QA9464 can choose the display of the CICS STID variable:
UCICSCNT MG110XN - IBM Description: 2:Storage Manager DSA
UTILVREF MG110XD - MXG Dataset Name 2:CICSMDSA
FORMATS MG110XL - MXG Dataset + Label 2:CICSMDSA:SMS STATS
VMXGINIT The existing default MG110XN format is unchanged.
Sep 18, 2019 In your SYSIN %LET UCICSFMT=MG110XD.; sets your choice.
-New format $MGDSNLAB maps MXG Dataset Name to the Dataset
Label for all MXG datasets.
Thanks to MP Welch, Bank of America, USA.
Change 37.206 BUILDPDB PDB.STEPS & PDB.JOBS are enhanced with variable
BUILD005 CPUZIPTM_CPUIFATM_INST to count zIIP instructions; the
BUIL3005 CP instruction count is already in variable CPU_INST.
Sep 13, 2019
Thanks to Jim S. Horne, Lowe's Companies, USA.
Change 37.205 TYPE71 LFAREA 1Mb FRAMES can be broken up into smaller
VMAC71 pages, and the number of BrokenUP frames is in the new
Sep 13, 2019 variable SMF71BRKUP='PGBL 1MB*BREAKUP*FRAMES', calculated
as the delta between installed 1MB Pageable frames and
the used 1MB frames.
Thanks to Joe Faska, DTCC, USA.
Thanks to Toni Skrajnar, IBM Support, USA.
Change 37.204 -z/OS 2.4 MXG RMF Monitor III support.
ADOCRMFV -New MXG Support for RMF III XCRG3 table:
ASMRMFV -New ASMRMFV Field Data Filtering (FDF) support for
VMACRMFV 4 more RMF III tables: CPDG3 ENTG3 GEIG3 OPDG3.
Sep 12, 2019 -15 ASMRMFV corrections for conditions found during
Sep 20, 2019
Sep 26, 2019
Oct 3, 2019 *** New Support ***
Oct 8, 2019
-Support for the RMF Monitor III Cryptographic Hardware
Data table (CRYG3) table new with z/OS 2.4 . The CRYG3
table does not exist for prior z/OS releases.
-The CRYG3 selection option is CRY (alias K).
The CRYG3 filtering option is NOCRY (aliases -CRY, -K).
CRYG3 is also selected if the MOST option is used.
-Support for the RMF Monitor III XCF Activity Data table
(XCFG3) which has existed at least since z/OS 1.13 but
was undocumented.
-The XCFG3 selection option is XCF (alias X).
The XCFG3 filtering option is NOXCF (aliases -XCF, -X).
XCFG3 is also selected if the MOST option is used.
-FDF support added for the RMF III Channel Data table
CPDG3. Information on CPD FDF filtering appears in
Section 33 Filtering The CPD Table in the ADOCRMFV
documentation member. There is a data dictionary listing
all the CPDG3 field names supported by FDF.
-FDF support added for the RMF III Enqueue Name table
ENTG3. Information on ENT FDF filtering appears in
Section 37 Filtering The ENT Table in the ADOCRMFV
documentation member. There is a data dictionary listing
all the ENTG3 field names supported by FDF.
-FDF support added for the RMF III OMVS Process Data table
OPDG3. Information on OPD FDF filtering appears in
Section 39 Filtering The OPD Table in the ADOCRMFV
documentation member. There is a data dictionary listing
all the OPDG3 field names supported by FDF.
-FDF support added for the RMF III General Information
table GEIG3. Information on GEI FDF filtering appears in
Section 38 Filtering The GEI Table in the ADOCRMFV
documentation member. There is a data dictionary listing
all the GEIG3 field names supported by FDF.
-Support for new release z/OS 2.4 RMF III fields added to
existing FDF Variable Name Tables.
*** Enhancements ***
-FDF support expanded for up to 256 byte character fields
raised from 8 bytes. But FDF IF expressions for
character field filtering may still not exceed the actual
size of the field or an error is flagged.
-Pattern matching subroutine MATCH now supports up to 256
characters in a pattern.
-FDF now translates X'00' to blank X'40' for selected
character field source data fields similar to what the
VMACRMFV SAS member does during a PDB build.
-Messages RMFV080I and RMFV088I now support character
string displays up to 105 characters. Longer character
strings are either output as a separate message or a '+'
flag is shown in the last byte of the message to indicate
the message has been truncated.
-Messages RMFV080I and RMFV088I now show tags of 'SFP' for
short floating point fields or 'LFP' for long floating
point fields instead of only showing 'FP'.
-Messages RMFV080I and RMFV088I now show the full IF
expression used if space allows.
-The SHOWBYTE subroutine now supports formatting of
storage byte values in kibibytes (kilobytes) 1024 through
yobibytes (yottabytes) 1024**8 in message RMFV088I.
However, current implementation restrictions at this time
limit values to a maximum of 9,223,372,036,854,775,807 or
about 1.22 pebibytes (petabytes).
-These documentation sections in member ADOCRMFV are all
updated for new support, enhancements, z/OS 2.4 and
corrections:
2 Terminology
3 Execution JCL
4 RMF III Table Selection Parameters
5 Input Data Selection Parameters
6 Report Control Parameters
7 Output Data Control Parameters
8 Error Handling Parameters
9 JCL and SYSIN Parameter Usage
12 Messages
13 Filtered Records
15 Program and IBM Limitations
21 Extended ASIG3/ENCG3/RCDG3/UWDG3 Record Support
22 RMF III VSAM Data Set Index Usage and Sizing
23 RMF III Options That Effect Data
24 RMF III Sysplex Master Gatherer
25 Ranges and Patterns
26 ASMRMFV and MXG PDB Data Relationships
31 Field Data Filtering (FDF)
32 Filtering The ASI Table
33 Filtering The CPD Table
34 Filtering The CSR Table
35 Filtering The DSI Table
36 Filtering The DVT Table
37 Filtering The ENT Table
38 Filtering The GEI Table
39 Filtering The OPD Table
40 Filtering The SPG Table
41 Filtering The SSH Table
42 Summary
43 Bibliography
*** Corrections ***
-All of these possible corrections apply to ASMRMFV in
MXG Versions 37.03-37.06.
-The last character of the 3 character day name in message
RMFV041I was truncated.
-Hex characters used in IF expressions for character
fields were not handled correctly.
-FDF IF expression numeric values with both a fraction and
exponent were not correctly scaled in subroutine IFNUM.
-Possible S0C7 Abend with large IF expression numeric
value and a Scale Multiplier in FDF IFNUM subroutine.
-Possible S0C7 Abend in FDF IFNUM subroutine processing
short or long floating point values.
-Possible Abend S0C4 in FDF SETIF subroutine because VNT
extension length was not added to base VNT entry length
to determine when IF entry array expansion was required.
-Incorrect formatting for a fixed binary number with a
fraction in FDF RMFV088I message.
-Incorrect digit shifting for fixed binary numbers in FDF
IF expression coded with a fractional part.
-Abend S0C4 in FDF SETIF subroutine when NOSHOWARR option
(default) was in effect and an IF array expansion
occurred.
-Possible S0C4 Abend in FDF SETIF subroutine when the
fieldname in an FDF IF expression was an alias.
-Incorrect formatting of some fractional numbers by
SHOWDEC subroutine in RMFV088I messages.
-Message RMFV088I after a FDF floating point precision
loss is detected incorrectly showed the original user IF
expression value instead of the true floating point value
actually in effect for comparisons.
-FDF subroutine IFNUM did not shift out any fractions from
the original user value in the IF expression to create an
integer when determining whether a floating point
precision loss had occurred.
-Possible incorrect byte display in FDF RMFV088I message
from SHOWBYTE subroutine for large values.
-Possible S0C1 Abend if a Fieldname entry in a FDF
Variable Name Table incorrectly has an alias equal to
itself.
*** FDF Limits ***
These are the current FDF supported limits for Change
37.204 as FDF IF expression values.
In no case may the IF expression value exceed the size of
the RMF III source field whether numeric or character.
Field Type Value range
---------- -------------------------------------------
Fixed Binary
1 byte 0-255
X'00'-X'FF'
2 byte 0-65,535
X'0000'-X'FFFF'
4 byte 0-4,294,967,295
X'00000000'-X'FFFFFFFF'
8 byte 0-9,223,372,036,854,775,807 *
X'0000000000000000'-X'7FFFFFFFFFFFFFFF' *
* the limit for a 8 byte binary number is imposed by the
need for ASMRMFV to convert the input number to binary.
A grande 64-bit register only holds X'7FFFFFFFFFFFFFFF'
as a high order sign bit is required.
Floating Point **
4 byte 0-9,223,372,036,854,775,807
8 byte 0-9,223,372,036,854,775,807
** the limit for a floating point number is imposed by
the need for ASMRMFV to convert the input number to
binary before conversion to floating point. A grande
64-bit register only holds X'7FFFFFFFFFFFFFFF' as a high
order sign bit is required.
Bit String
1 byte 0-255
X'00'-X'FF' without don't cares
1 byte .......0-1111111. with don't cares
Percentage 0.0-100.0
Time of Day 01JAN2000:00:00:00-17SEP2042:23:53:46 ***
*** The of day limit is imposed by the 64 bit TOD clock
which wraps to all binary ones in September 2042. The
addition of leap seconds may affect this value.
Time ****
4 byte 0-4,294,967,295
X'00000000'-X'FFFFFFFF'
8 byte 0-9,223,372,036,854,775,807
X'0000000000000000'-X'7FFFFFFFFFFFFFFF'
**** Time units depend on the default time unit for the
RMF III field or the explicit time unit coded by the
user. They can range from microseconds to days.
TOD date only 01JAN2000-17SEP2042
TOD Time only 00:00:00-23:59:59
Change 37.203 Unused Change Number.
Sep 11, 2019
Change 37.202 Support for truncated SMF 61 catalog record from z/OS 2.3
VMAC6156 causing INPUT STATEMENT EXCEEDED error.
Sep 9, 2019
Thanks to Mike Jacques, BB&T, USA.
Change 37.201 The MGBYTES format is extended to decode storage units in
FORMATS zettabytes and yottabytes.
Sep 27, 2019
Change 37.200 Optional CICSTRAN DBCTL segment variable STATCTM1 was
IMACICDB 10 times too large (4 usec should be 0.4 usec) due to
Sep 9, 2019 incorrect STCK informat, corrected to TU4.
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.199 -Variables QWACPCTT/QLACRLNU are used to get the correct
VMACDB2 count of THREADS when ROLLUPS (DB2PARTY=R) are in use:
ASUMDB2A DB2ACCT variable QWACPCTT is summed into THREADS in
ASUMDB2B ASUMDB2A unless QLACRLNU is GT 0 and used instead.
ASUMDB2G DB2ACCTP variable QPACRLNU is summed into PACKCNT in
ASUMDB2P ASUMDB2P
ASUMDB2R DB2ACCTB variable QWACPCTT is summed into THREADS in
Sep 7, 2019 ASUMDB2B unless QLACRLNU is GT 0 and used instead.
Sep 16, 2019 DB2ACCTG variable QLACPCTT is summed into THREADS in
ASUMDB2G unless QLACRLNU is GT 0 and used instead.
DB2ACCTR variable QLACRLNU is summed into THREADS
-Previously THREADS in DB2ACCT and COUNT in DB2ACCTP
counted each record; so counts will be larger and
correct in the ASUMs and in ANALDB2R which uses them,
unless the count is one, which does happen.
-This change means that the average values reported in
ANALDB2R will be more accurate but it still precludes
detail problem analysis from rolled up records since
the best you can get is an average value.
Variable QWACPCTT is now kept in DB2ACCTB/DB2ACCTG/
DB2ACCTR so it can be used in those ASUMs.
Change 37.198 BETA 93 Subtypes 12,17,30,31,55 were not input because
VMACBETA they weren't added to the IF SUBTYPE IN list of subtypes,
Sep 11, 2019 and they were not sorted to the PDB data library.
Thanks to Andreas Menne, Finanz Informatik, GERMANY.
Change 37.197 BETA 97 Subtype 51 Dataset BETA9751D did not input the
VMACBE97 "New Area" fields for BETA9751REC='U', UPDATE records.
Sep 11, 2019 An undocumented test to skip U records was removed.
Thanks to Andreas Menne, Finanz Informatik, GERMANY.
Change 37.196 Member IMACTIME and CHANGESS had references to "OKJOB"
IMACTIME that should be OKFLAG for consistency; no impact as the
CHANGESS member is optional and "OKJOB" is in comments.
Sep 5, 2019
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.195 The PDB.PRINT dataset ACCOUNT variables were populated
BUILD005 only in the first observation for each job, and TYPETASK
Sep 5, 2019 had the SUBSYS6 value (PSF/VPS) instead of TYPETASK (JOB)
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.194 Formats added for TYPE42DS variables S42DSENT/S42DSCMT to
FORMATS decode the Encryption Type and Compression TYPEs.
VMAC42 ET=AES-256,CT=None/Generic/Tailored/zEDC.
Sep 3, 2019
Thanks to Luis Mendoza, BKFS, USA.
Change 37.193 Format $MG110EX for dataset CICSEXCE variable EXCMNTYP
FORMATS did not map value '0004'X to 04:POLICY THRESHOLD instead
Sep 3, 2019 printing a confusing '00'X value.
Thanks to Harald Seifert, Huk-Coburg, GERMANY.
Change 37.192 TLMS dataset TYPETLMS from the "B" record is updated and
EXTYTLMC two new TLMS datasets TYPETLMC/TYPETLMD are now created
EXTYTLMD from the "C" and "D" TLMS records:
IMACTLMS DDDDDD DATASET DESCRIPTION
VMACTLMS TYTLMC TYPETLMC MASTER FILE CONTROL RECORD
VMXGINIT TYTLMD TYPETLMD VOLUME MASTER FILE MULTI DATASET
Sep 25, 2019
Thanks to Pierre Pascal Joulin, SOCGEN, FRANCE.
Change 37.191 RMM/EDGR processing in VMXGDSN had zero obs for TAPES and
VMXGDSN TAPEDSNS MXG 37.03-MXG 37.06; the SORT input dataset was
Sep 2, 2019 incorrectly changed to EDGRDEXT instead of EDGRXEXT.
Thanks to Wayne Bell, UNIGROUP, USA.
Thanks to John Fulton, UNIGROUUP,USA
====== CHANGES THRU 37.190 ARE IN MXG 37.06 DATED AUG 30, 2019 =========
Change 37.190 New macro %MXGFINFO creates dataset EXTFILES with these
MXGFINFO variables for every external filename (INFILE)
Aug 30, 2019 -z/OS: FILEREF DSNAME DEVICE and CREATEDATE
-ASCII FILEREF XPATH and CREATEDATE.
CREATEDATE will be missing if the ftp access method is
used.
Change 37.189 Spurious INVALID VALUE FOR INPUT FUNCTION in first 37.06
DODSCRDT had no impact, except for lots of lines on the log. It
Aug 30, 2019 occurs when the INFILE is on TAPE, because the DSCB that
SAS returns is the VOL2HDR instead of the date-containing
DSCB, so the CREATEDATE variable is always missing for
INFILE on Tape.
Thanks to Jim S. Horne, Lowe's Companies, USA.
Change 37.188 Example 2 had a Missing "END;" statement
IMACUOW
Aug 29, 2019
Change 37.187 Labels for Tennant TRG_SUCP,TRG_SUIFA,TRG_SUSP variables
VMAC7072 are changed from *MSU* to *HDW MSU* because those values
Aug 29, 2019 are NOT the Software MSU (4HR AV) we normally use when
discussing MSU. This link shows IBM uses the "Hardware"
SU_SEC value to convert those service units to engine
counts:
https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/
com.ibm.zos.v2r3.izsc100/cserbb200195.htm
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.186 Support for IBM Tivoli Advanced Allocation Management SMF
EXTYAAM1 record; this product replaces the old X37 product.
EXTYAAM3 DDDDDD DATASET DESCRIPTION
FORMATS TYAAM1 TYPEAAM1 SUBSYSTEM ACTIVITY
IMACAAM TYAAM3 TYPEAAM3 PROCESSING ACTIVITY
TYPEAAM Subtype 3 have been data tested; IBM provided additional
TYPSAAM bit values for DNV/DST/ADP and reported SPCF incorrectly
VMACAAM sets '80'x bit causing 'C0'x for TRACKS, to be corrected,
VMXGINIT but MXG's format maps both 'C0'x and '40'x to TRACKS.
Aug 28, 2019 Subtype 3 event records have only SMFTIME & JOB (AAMJBN),
Sep 24, 2019 no READTIME nor JCTJOBID/JESNR, so they can not be easily
interleaved/merged with other JOB-related records.
Thanks to Cha Kihun, Navy Federal, USA.
Thanks to Richard Champouillon, Navy Federal,USA
Change 37.185 Warning: APPARENT SYMBOLIC REFERENCE LDB and LDB@ACG has
READDB2 no impact on the results, DB2ACCTG is correctly sent to
Aug 23, 2019 your LDB2ACG= argument with correct message text.
Thanks to Douglas C. Walter, CITIGROUP,USA.
====== CHANGES THRU 37.184 ARE IN MXG 37.06 DATED AUG 22, 2019 =========
Change 37.184 Delete of temp dataset SRTIRC was relocated so it is
VMXGCICI always deleted (to free WORK space).
Aug 14, 2019
Change 37.183 Support for SMF 82 subtypes 30 and 40-48 new datasets
FORMATS DDDDDD DATASET DESCRIPTION
EXTY82AU TY8230 TYPE8230 KDS ARCHIVE/CRYPTOPERIOD
EXTY8248 TY8240 TYPE8240 CCA SYMMETRIC KEY LIFECYCLE
IMAC82 TY8241 TYPE8241 CCA ASYMMETRIC KEY LIFECYCLE
VMAC82 TY8242 TYPE8242 PKCS#11 KEY LIFECYCLE EVENT
VMXGINIT TY8243 TYPE8243 RCS CONFIGURATION CHANGE (not decoded)
Aug 21, 2019 TY8244 TYPE8244 CKDS KEY USAGE
Aug 27, 2019 TY8245 TYPE8245 PKDS KEY USAGE
Oct 30, 2019 TY8246 TYPE8246 PKCS#11 KEY USAGE
TY8247 TYPE8247 PKCS#11 NOKEY USAGE
TY8248 TYPE8248 WARN MODE
Subtype 43 is not decoded, pending test data records.
Subtypes 40,41,44,45,48 have been tested.
Subtypes 42,46,47 are decoded but not tested.
Thanks to Alexander Bitter, Worldpay, USA.
Thanks to Lethika Panicker, Worldpay, USA.
Thanks to Ron Rust, Worldpay, USA.
Change 37.182 Variable SMF74SCMR was incorrectly spelled SMF74SKCR.
VMAC74
Aug 14, 2019
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.181 Support for Hitachi MAR Mainframe Analytics Recorder 9.1
EXMAR07 creates new MARST07 PARM Section Dataset.
IMACMAR
VMACMAR
VMXGINIT
Aug 13, 2019
Change 37.180 MXG DB2 Formats $MGTMDOB, $MGTMDRM, $MGTMDRE, MGTMDRC are
FORMATS updated with new values.
Aug 13, 2019
Thanks to Randy Hewitt, DXC, USA.
Change 37.179 TYPE72GO variables METGOAL and PCTMETGO were wrong, now:
VMAC7072 IF TRANS GT 0 THEN DO M=1 TO 6;
Aug 9, 2019 METGOAL=SUM(METGOAL,RTSTRN(M));
PCTMETGO=100*METGOAL/TRANS;
END;
ELSE PCTMETGO=.;
Thanks to James Peddycord, Northern Trust, USA.
Thanks to Karl S. Huf, Northern Trust, USA.
Thanks to Arati Khodaskar, IBM Global Services, USA.
Change 37.178 -MXG 37.05 only, possible S0C4 Abend in ASMRMFV PROCSSH
ASMRMFV subroutine when comparing current RMF Version to a z/OS
Aug 9, 2019 2.3 Version number, after Change 37.140.
Aug 20, 2019 -That PROCSSH code was redundant with similar code in the
FINDPOL subroutine that is RMF Version independent and
has been removed to eliminate the S0C4 possibility.
-MXG 37.03-37.05, ONLY if you use FDF IF expressions.
ABEND S0C7 when a hex value is coded for a numeric field
in an FDF IF expression. For example: IF=(ASIDP EQ X'FF')
-Incorrect handling of exponents when coded for a numeric
field in an FDF IF expression causing an incorrect
compare value to be calculated.
For example: IF=(ASI1MBFF GT 1E2)
-Invalid hex value can be shown in RMFV080I and RMFV088I
messages.
Thanks to Kurt Gramling, GTS Tech-Support: CRM, USA
Change 37.177 Variables R745BYTR/BYTW/RTIR/RTIW were never populated by
VMAC74 IBM, as they were replaced by R7451CT1-R7451CT4, but they
Aug 8, 2019 are now populated by those replacement values rather than
being missing values. See Change 23.314.
Thanks to Otto Burgess, OPM, USA.
Change 37.176 Support for IMS LOG '02'x record creates IMS02 dataset or
EXIMS02 prints a message if a multi-segment command record is
VMACIMS found to send your IMSLOG so it can be supported.
VMXGINIT
Aug 7, 2019
Change 37.175 New metric, SIISPCT='STORE INTO*INSTRUCTION*STREAM*PCT'
ANALSIIS is added to TYPE1131 and ASUM1131 datasets, to identify
VMAC113 potential timeframes based on percent of certain I writes
Aug 6, 2019 vs D Writes sourced, to identify when it happens, but NOT
who is causing it.
-ANALSIIS identifies intervals with SIISPCT GT 10 percent
and identifies what programs were running in descending
CPUTM or CPUZIPTM depending on CPU type during that high
SIISPCT interval.
Thanks to Kathy Walsh, IBM zSystems, USA.
Thanks to John Burg, IBM zSystems, USA.
Change 37.174 Non-fatal Divide By Zero when QBSTVPL=0 was corrected.
VMACDB2
Aug 5, 2019
Ron van der Zande, KLM Information Systems, THE NETHERLANDS.
Change 37.173 Support for TPMX $JCL_JJR, variable JCLJJR in TYPETPMX.
VMACTPMX Only 9 lines of "new field" messages are now printed.
Aug 4, 2019
Thanks to Jack Hyde, Optum Technology, USA.
Change 37.172 Dataset Encryption Variable SMF14DEF='Y' identifies data
VMAC1415 sets that are encrypted, and SMF14DET='0100'x to indicate
Aug 4, 2019 AES ('01'x) and 256 Bits ('00'x). INPUT was corrected.
Change 37.171 Variables R723GGTI, R723GGTN, R723GGKY and R723MFLG are
VMAC7072 kept in dataset TYPE72GO, and variable R723GGKY is now
Aug 4, 2019 INPUT correctly as $EBCDIC64 instead of 32.
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.170 Variable DESTNATN is added to PDB.TYPE6 dataset.
BUILD005
BUIL3005
Aug 2, 2019
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.169 Variable INITTIME in dataset TYPE30_6 is created using
VMAC30 INITTIME=SMFTIME-ACTIVETM;
Aug 2, 2019 which can then be used to count unique step executions.
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.168 Four CICS "identity" variables weren't kept in CICSTRAN
UTILEXCL when UTILEXCL was used: APPLID JOB SMFPSSPN SMFPSRVR, and
VMAC110 these new "identity" variables MVSLEVEL LOCLINFO MCTSSCRL
Aug 2, 2019 are added to CICSTRAN whether UTILEXCL is used or not.
All could be deleted with
%LET MACKEEP=
MACRO _KCICTRN
DROP=APPLID JOB SMFPSSPN SMFPSRVR
MVSLEVEL LOCLINFO MCTSSCRL
%
:
in your SYSIN.
Thanks to Scott Barry, SBBWorks, INC., USA.
Change 37.167 z/OS 2/4 updates for RMF MONITOR III:
VMACRMFV -VMACRMFV in MXG 37.05 (only) fails with z/OS 4.2 CSR data
Aug 2, 2019 and was corrected in 37.06 by this change (which changed
only line 6871, from /48 to /CSRENTLE).
-New RUCSA variables added to ZRBCSR dataset:
CSRRUCSA ='RUCSA*AMOUNT'
CSRERUCSA='ERUCSA*AMOUNT'
-New RUCSA variables added to ZRBGEI dataset:
GEIRUCSASZ='IPL SIZE*RU CSA*BELOW 16MB'
GEIERUCSAZ='IPL SIZE*RU CSA*ABOVE 16MB'
GEIRUCSAMX='MAX RUCSA*BELOW 16MB'
GEIERUCSAX='MAX RUCSA*ABOVE 16MB'
GEIRUCSASP='ALLOCATED RUCSA*BELOW 16MB'
GEIERUCSAP='ALLOCATED RUCSA*ABOVE 16MB'
GEIRUCSAAV='ACCUM RUCSA*BELOW 16MB'
GEIERUCSAV='ACCUM RUCSA*ABOVE 16MB'
GEIRUCSARE='UNALLOCATED*RUCSA BELOW 16MB'
GEIRUCSAAS='ACCUM RUCSA*BELOW 16MB*BY SYSTEM'
GEIERUCSAS='ACCUM RUCSA*ABOVE 16MB*BY SYSTEM'
GEIBATRUCSA='ACCUM RUCSA*BELOW 16MB*BY BATCH'
GEIBATERUCSA='ACCUM RUCSA*ABOVE 16MB*BY BATCH'
GEIASCRUCSA='ACCUM RUCSA*BELOW 16MB*BY ASCH'
GEIASCERUCSA='ACCUM RUCSA*ABOVE 16MB*BY ASCH'
GEIOMVRUCSA='ACCUM RUCSA*BELOW 16MB*BY OMVS'
GEIOMVERUCSA='ACCUM RUCSA*ABOVE 16MB*BY OMVS'
-New variables with time to LPAR/Group Capping in ZRBCPU:
CPC_TIME_TO_CAPL='TIME TO*LPAR*CAPPING'
CPC_TIME_TO_CAPG='TIME TO*GROUP*CAPPING'
Thanks to Kurt Gramling, TSYS, USA.
Change 37.166 z/OS 2/4 updates:
VMAC74 -Type 74 Subtype 2 additions to dataset TYPE74PA, virtual
VMAC78 storage for the optional Private Address Space data:
Aug 2, 2019 R742PUTM1 ='PATH1*TIME USED*AT PCT UTIL'
R742PUTMS1='PATH1*SQRD TIME USED*AT PCT UTIL'
R742PUCN1 ='PATH1*COUNT USED*AT PCT UTIL'
R742PUSCN1='PATH1*SIGNAL COUNT SENT'
R742PUPCT1='PATH1*PCT UTIL'
R742PUTM2 ='PATH2*TIME USED*AT PCT UTIL'
R742PUTMS2='PATH2*SQRD TIME USED*AT PCT UTIL'
R742PUCN2 ='PATH2*COUNT USED*AT PCT UTIL'
R742PUSCN2='PATH2*SIGNAL COUNT SENT'
R742PUPCT2='PATH2*PCT UTIL'
R742PUTM3 ='PATH3*TIME USED*AT PCT UTIL'
R742PUTMS3='PATH3*SQRD TIME USED*AT PCT UTIL'
R742PUCN3 ='PATH3*COUNT USED*AT PCT UTIL'
R742PUSCN3='PATH3*SIGNAL COUNT SENT'
R742PUPCT3='PATH3*PCT UTIL'
R742PUTM4 ='PATH4*TIME USED*AT PCT UTIL'
R742PUTMS4='PATH4*SQRD TIME USED*AT PCT UTIL'
R742PUCN4 ='PATH4*COUNT USED*AT PCT UTIL'
R742PUSCN4='PATH4*SIGNAL COUNT SENT'
R742PUPCT4='PATH4*PCT UTIL'
R742PNIBTM='PATH TOTAL TIME*NO INBOUND*BUFFER IMPACT'
R742PNIBTS='PATH SQUARED TIME*NO INBOUND*BUFFER IMPACT'
R742PNIBCN='PATH COUNT*NO INBOUND BUFFER'
/* XCFMGD */
-Type 78 subtype 2 dataset TYPE78VS Virtual Storage new:
R782RUCA ='RUCSA ADDRESS*BELOW 16MB'
R782RUCS ='RUCSA SIZE*BELOW 16MB'
R782ERUCA='RUCSA ADDRESS*ABOVE 16MB'
R782ERUCS='RUCSA SIZE*ABOVE 16MB'
Thanks to Kurt Gramling, TSYS, USA.
Change 37.165 Dataset TYPE8201 (Initialization) variables SMF82ITE/CKD/
VMAC82 IML/USR/PKD were misaligned by a one byte reserved field.
Jul 29, 2019
Thanks to Matthew T Chappel,CQueensland Dept Transport, AUSTRALIA
Change 37.164 Variable TTTTLSSP in dataset TYP11902 is decoded by new
FORMATS $MG119PT format:.
VMAC119 VALUE $MG119PT /*TTTTLSPP*/
Jul 28, 2019 '0200'X='0200X:SSL V2'
'0300'X='0300X:SSL V3'
'0301'X='0301X:TLS 1.0'
'0302'X='0302X:TLS 1.1'
'0303'X='0303X:TLS 1.2'
Variable TTTTLSNC documents '4X'x='USE TTTTLSNC4 instead'
but values of '0A'x '35'x and '6B'x are found in data
but are not documented.
Thanks to Joe Faska, DTCC, USA.
Change 37.163 Labels for these variables were made consistent
VMAC71
Aug 2, 2019 SMF71L4A='AVG*1MB*PAGEABLE*FRAMES*IN DREF'
SMF71L4M='MIN*1MB*PAGEABLE*FRAMES*IN DREF'
SMF71L4X='MAX*1MB*PAGEABLE*FRAMES*IN DREF'
SMF71L5A='AVG*1MB*AVAILABLE*FRAMES*IN DREF'
SMF71L5M='MIN*1MB*AVAILABLE*FRAMES*IN DREF'
SMF71L5X='MAX*1MB*AVAILABLE*FRAMES*IN DREF'
SMF71L6A='AVG*1MB*PAGEABLE*FRAMES*USED*IN DREF'
SMF71L6M='MIN*1MB*PAGEABLE*FRAMES*USED*IN DREF'
SMF71L6X='MAX*1MB*PAGEABLE*FRAMES*USED*IN DREF'
SMF71L8A='AVG 1MB*PAGEABLE*FRAMES*IN CSTORE'
SMF71L8M='MIN 1MB*PAGEABLE*FRAMES*IN CSTORE'
SMF71L8X='MAX 1MB*PAGEABLE*FRAMES*IN CSTORE'
SMF71L9A='AVG 1MB*AVAILABLE*FRAMES*IN CSTORE'
SMF71L9M='MIN 1MB*AVAILABLE*FRAMES*IN CSTORE'
SMF71L9X='MAX 1MB*AVAILABLE*FRAMES*IN CSTORE'
SMF71PLA='AVG*1MB*PAGEABLE*FRAMES*USED*IN CSTORE'
SMF71PLM='MIN*1MB*PAGEABLE*FRAMES*USED*IN CSTORE'
SMF71PLX='MAX*1MB*PAGEABLE*FRAMES*USED*IN CSTORE'
Aug 2, 2019
Thanks to Joe Faska, DTCC, USA.
Change 37.162 ANALSIZE failed due to a missing semicolon in VMXGSIZE.
VMXGSIZE
Jul 25, 2019
Thanks to Richard Haynes, BCBSKS, USA.
Change 37.161 -New z/OS-Only SAS Date Variable CREATEDATE can be created
DODSCRDT for INFILE names of SAS CONTROLT DCOLLECT RMFBSAM TMC
VMACCTLT EDGHSKP IMSLOG OPCLOG, and is that variable is available
VMACDCOL the IHDRxxxx exit for your selection criteria, and/or it
VMACEDGR can be kept using the _Kdddddd dataset KEEP macro.
IHDRTMS5 -You must enable its creation with %LET DSCRDT=YES; in the
VMACIMS SYSIN or %DSCRDT can be added for any z/OS INFILE by
VMACSMF adding JFCB=MXGJFCB DSCB=MXGDSCB to the INFILE statement
VMACRMFV and inserting %DSCRDT after the first INPUT statement:
VMACTMS5 DATA MYSTUFF;
VMXGINIT INFILE MYFILE JFCB=MXGJFCB DSCB=MXGDSCB;
Aug 7, 2019 INPUT ;
Aug 11, 2019 %DSCRDT(JFCB=MXGJFCB);
Nov 26, 2019 SAS automatically prints the Create Date on the SAS log
in its "INFILE IS " message, but DSCRDT can print its
own log message if you use %LET MXGEXIMSG=YES;
SEE CHANGE 37.249 FOR REVISION.
Thanks to Linda S. Berkley, DISA, USA
Change 37.160 Unused Change Number.
Change 37.159 BETA 93 610 Subtype 40 and 49 misaligned, INVALID DATA
VMACBETA messages for variables BETASTME and BETAETME in 49 and
Jul 18, 2019 BETAALT in subtype 40.
Change 37.158 BLDSMPDB adds SPIN: SPUN: to WEEKDROP MNTHDROP if they
BLDSMPDB are not present. New parameters added WEEKBASE MNTHBASE
Jul 15, 2019 that both default to blanks which will then become
yesterday and used to determine which datasets will be
included in the weekly and monthly PDBs. This ensures
that your most current PDB is used to build the
weekly/monthly datasets. So if you decided to add
something on the day before it will be propagated into
the weekly/monthly jobs. The fact that it may not exist
in all of the input PDBs is not a problem. These should
only be used if you wish to force a specific PDB to be
the basis for the weekly/monthly PDBs and if it does not
exist via either a LIBNAME or a DD the job will fail.
Change 37.157 A new macro that will drill down through MSU consumption
ANALMSUS from the TYPE89 records and the type 30 interval data. It
Jul 13, 2019 can produce bar charts, tabular reports, and EXCEL
Aug 27, 2019 spreadsheets as you choose.
Change 37.156 Variables ABEND CONDCODE added to PDB.SMFINTRV and the
VMAC30 TYPE30_V datasets. The values will only be populated from
Jul 13, 2019 the subtype 3 records.
Thanks to Scott Barry, SBBWorks Inc., USA.
Change 37.155 Support for CICS Optional field USER/AGENCY.
IMACAAAA
IMACICXB
UTILEXCL
VMAC110
Thanks to Mark Hiltbruner, State of South Dakota, USA.
====== CHANGES THRU 37.154 ARE IN MXG 37.05 DATED Jul 8, 2019 =========
Change 37.154 SMF 110 Subtype 1 MNSEGCL=5 INPUT EXCEEDED ERROR due to
VMAC110 8 byte reserved field inserted in DPL ENTRY segment. Skip
Jul 8, 2019 the records (causing zero obs in CICSRDPL dataset) with:
%LET MAC110H= %QUOTE(IF MNSEGCL=5 THEN DELETE; ) ;
This was added in CICS/TS 5.5 to these (seldom used) CICS
Resource segment.
Thanks to Jack Hyde, Optum Technology, USA.
====== CHANGES THRU 37.153 ARE IN MXG 37.05 DATED Jul 6, 2019 =========
Change 37.153 SMF 120 Subtype 3 INPUT STATEMENT EXCEEDED ERROR due to
VMAC120 incorrect MXG logic that has accidentally worked: there
Jul 6, 2019 were only three ABENDS since Dec 2018. One circumvention
which usually skips over the failing record was tried:
OPTIONS STOPOVER MISSOVER % but because the code error
was a loop on the same INPUT location, MXG created 500
million observations in TYP120SR filling fifteen WORK
volumes before dying with a B37 no more extents error.
The alternative circumvention was to skip that subtype:
%LET MACFILE=%QUOTE(IF ID=120 AND SUBTYPE=3;);
until this update corrected the MXG code error.
I think not related, but site had WebSphere 8.5.5 FP12.
====== CHANGES THRU 37.152 ARE IN MXG 37.05 DATED Jul 5, 2019 =========
Change 37.152 MXGSTEP populates new variable MXGSTEP='Y' in SMF 30's to
MXGSTEP identify job steps that execute MXG programs, populating
Jul 5, 2019 TYPE30_V and TYPE30_4 with (TYPE30) and in PDB.SMFINTRV
and PDB.STEPS with (BUILDPDB), or any MXG program that
processes SMF 30 records, if PROGRAM='SAS' and DDNAMES
SOURCLIB and LIBRARY are in this STEP, as both are
required for MXG Execution.
Thanks to Deepa Rajendran, DXC, SINGAPORE.
Change 37.151 ASUMMIPS now uses the $MGRMIPS format built from the IBM
ASUMMIPS LSPRITR table to lookup CPCFNAME (eg 3906-716) for the
VMXGINIT MIPSFACT (eg 8.34), the MIPS per MSU. Previously you
Jul 4, 2019 had to provide your own MIPSFACT.
Jul 5, 2019
Thanks to Randy Hewitt, DXC, USA.
Change 37.150 Support for DATACOM log file.
EXDCOM
IMACDCOM
TYPEDCOM
TYPSDCOM
VMACDCOM
VMXGINIT
Jul 2, 2019
Thanks to Linda S. Berkley, DISA, USA.
Change 37.149 If you add ID to USERADD it must be the last entry
UTILBLDP in the list. If you happened to make it first the
Jul 2, 2019 list will be adjusted.
Change 37.148 BLDSMPDB adds SPIN: SPUN: to WEEKDROP MNTHDROP if they
BLDSMPDB are not present. New parameters added WEEKBASE MNTHBASE
Jun 23, 2019 that both default to &WEEKDATE (yesterday) are used to
decide which datasets can be included in the weekly and
monthly PDBs. This ensures that your most current PDB is
used to build the weekly/monthly datasets. So if you
decided to add something on the day before it will be
propagated into the weekly/monthly jobs. The fact that it
may not exist in all of the input PDBs is not a problem.
Change 37.147 CICS Statistics datasets CICMPR and CICSJN were not in
VMAC110 the _N110_, _S110, and _S110ST optional tailoring macros.
Jul 1, 2019
Change 37.146 Macro variable &MACSPIN added to IMACSPIN for "instream"
IMACSPIN tailoring of SPINCNT.
VMXGINIT
Jul 1, 2019
Thanks to Scott Barry, SBBWorks Inc., USA.
Change 37.145 The "PROC PRINT" output with Label and Name column heads
VMXGPRA1 VMXGPRA1 and VMXGPRAL utilities now protect if you have
VMXGPRAL changed the OBS option. Their temp dataset has one obs
Jun 28, 2019 per variable, but if your OBS was too small, some heads
were wrong and missing parens. Now the OBS value is held.
the print is accomplished and your original OBS restored.
Thanks to Scott Wiig, US Bank, USA.
Change 37.144 FLASH: MISSING PERIODS 2/3 TYPE72GO if MXG 36.07 or prior
VMAC7072 is used and IBM RMF APARs for SCM and Crypto are applied.
Jul 1, 2019
There is no error with MXG 36.08 (Sept 2018) or later.
One z/OS 2.3 site reported these were applied:
UA98434 APAR OA56461
UA98529 APAR OA56672
UA98759 APAR OA56826
UA98999 APAR OA56860
and one z/os 2.2 site reported this was applied
UA98433 APAR OA56747
but there may be other maintenance involved.
You can examine your SMF 72 Subtype 3 period data with
PROC FREQ DATA=PDB.TYPE72GO;
TABLES SYSTEM*PERIOD;
TITLE TABLE OF PERIOD VALUES IN SMF 72 SUBTYPE 3;
to see if you are missing values for your periods.
These missing periods will cause the system Capture
Ratio to decrease significantly, and the workload that
normally have period data will have reduced CPU time in
TYPE72GO and RMFINTRV datasets, and reports from them.
Change 37.143 Expanded Storage doesn't exist in all z/OS systems so the
IMAC71 56 ESTORE variables in TYPE71 can be dropped by removing
Jun 28, 2019 the comment block when you EDIT the IMAC71 tailoring
member into your "USERID.SOURCLIB" tailoring library.
You do need to examine any reporting programs that use
the TYPE71 dataset to see if those variables are used.
Thanks to Arnold Kim, UPS, USA.
Change 37.142 New READRATE %MACRO will measure the Read Rate (MiB/Sec)
READRATE of MXG processing SMF data records, printing an interval
VMACSMF trace on the log, a PLOT of READRATE vs RUNTIME, and PROC
VMXGINIT TABULATE report with 1 sec default interval. Additional
options are in the comments in READRATE. Syntax:
Jul 1, 2019 %READRATE(READRATE=1,RESULTS=BOTH);
%INCLUDE SOURCLIB(TYPE30); RUN;
&READRATEREPORT;
Change 37.141 Format MG119CD 17:UCP corrected to 17:UDP.
FORMATS
Jun 26, 2019
Thanks to Jenny Chen, DXC Technology, AUSTRALIA.
Change 37.140 -More RMF Monitor III tables are supported by FDF (Field
ADOCRMFV Data Filter) in ASMRMFV: DSI, SPG, SSH
ASMRMFV -New CDF (Character Data Filters) added for RMF III VSAM
Jun 26, 2019 data set level filtering:
CPCNAME= (aliases CPC=, CECNAME=, CEC=)
LPARNAME= (alias LPAR=)
-LPAR names and CPC names are validated for correct syntax
when these filters are used.
-FDF now supports character patterns for character fields.
Only Equal (= EQ) and Not Equal (^= =^ NE NEQ NOT NOTEQ
NOT=) operators may be used in an IF expression with a
character pattern. A pattern either matches or it does
not. Other FDF operators are flagged as an error.
-New DSIAND/DSIOR parameters which have the same function
as the prior SYSAND/SYSOR parameters which are now
respective aliases.
-RMFV014I message now shows counts for RMF III data sets
bypassed by CPCNAME= and/or LPARNAME= CDF keywords.
-Duplicate counts are no longer shown in RMFV014I message
if DUPDSN option is in effect (no duplicate checking).
-Space analysis messages RMFV030I, RMFV031I are no longer
issued for filtered RMF III VSAM data sets.
-MAXDSNS= added as a further alias of MAXDSNAMES=.
-Always force upper case for these CDF keyword values
because lower case letters are always invalid and
would be flagged as an error otherwise:
CPCNAME= SYSPLEX= LPARNAME= SYSTEM=
ASISUBSYS= ASIJOBCLASS= ASIJOBNAME= ASIJESID=
CSRJOBNAME= CSRJESID= DVTDEVNUM=
OPDJOBNAME= OPDPROCNAME= OPDUSERNAME=
-Support validation for all characters allowed for CDF
Workload Names, Service Classes, Report Classes, and
Resource Groups
-Field descriptions in data dictionary entries in ADOCRMFV
for FDF supported RMF III table expanded for better
clarity.
-TRUENAME Fieldnames in FDF data dictionary entries in
ADOCRMFV documentation now show all possible aliases.
-GMT offset value in Summary First Sample Begin Date/Time
selected message RMFV013I could be incorrect.
-RMFV013I selection messages were incorrectly displayed in
Summary report when all RMF III data sets were filtered.
-CDF keywords and aliases may now be used as Fieldnames
in FDF IF expressions (minus the = suffix).
-Message RMFV014I now includes a counter for FDF filters.
-Negative values are now supported in FDF, but only for
GMT offset fields xxxSTDIF and xxxGMTOFF where xxx is a 3
character RMF table id.
-For example this is a valid IF expression:
IF=(ASISTDIF EQ -5H)
-New second RMFV103I message is added to Detail and
Summary reports to display Sample Set filter reason
counts.
-New options AUTOSEL (alias AUTO) and NOAUTOSEL (alias
NOAUTO) added.
-AUTOSEL is the default and will result in the RMF III
table being automatically selected with any CDF or FDF
filter if the table was not already selected. This is a
convenience feature.
-New message RMFV082I appears when a table is auto
selected. In addition in message RMFV105I Y* will
appear in the SELECT column for auto selected tables.
-NOAUTOSEL provides the prior ASMRMFV behavior and the
unselected table condition generates an error. However,
it may be helpful if JCL with CDF and/or FDF filters is
routinely reused to avoid the PDB build overhead of
automatically selecting a table that is no longer wanted.
-For tables not referenced by CDF and/or FDF it is still
necessary to select the RMF III tables of interest.
-The SPGVOLSER= CDF filter could have incorrect results.
-Following documentation sections in ADOCRMFV are added or
updated:
Section Description
0 Contents
2 Terminology
3 Execution JCL
4 RMF III Table Selection Parameters
5 Input Data Selection Parameters
6 Report Control Parameters
8 Error Handling Parameters
9 JCL and SYSIN Parameter Usage
12 Messages
13 Filtered Records
16 Return Codes
20 FREE=CLOSE For VSAM Data Sets
21 Extended ASI/ENC/RCD/UWD Record Support
25 Ranges and Patterns
26 ASMRMFV and MXG PDB Data Relationships
31 Field Data Filtering (FDF)
32 Filtering The ASI Table
33 Filtering The CSR Table
34 Filtering The DSI Table
35 Filtering The DVT Table
36 Filtering The ENT Table (Future)
37 Filtering The GEI Table (Future)
38 Filtering The OPD Table (Future)
39 Filtering The SPG Table
40 Filtering The SSH Table
41 Summary
Change 37.139 Reserved Change.
Change 37.138 The label for variable QW0199TRS in DB2 102 IFCID 199 was
VMAC102 corrected to 'END TIME*OF*INTERVAL', which is strange as
Jun 24, 2019 the SMFTIME was available for the end time.
Thanks to Xing Su, DXC Technology, AUSTRALIA.
Thanks to Peter J. Gray, DXC Technology, AUSTRALIA.
Change 37.137 -Variables INDXUSEP and POLYUSEP percentages are created
VMACRMFV in dataset ZRBDISH to track index usage. The 1110 is the
Jun 21, 2019 maximum number of sample indexes in a 32K DSI table and
50 is the maximum number of policy indexes.
-Strange RMF III intervals can be created if the values
in SMFPRMxx don't match ERBRMFxx options. A site had
SMF INTVAL(10) SYNCVAL(59) with RMF MINTIME(300) SYNC(0)
for RMF III, which created a 4 minute interval (:55 -:59)
when the RMF III MINTIME expired, a one minute (:59-:00)
interval when the SMF Interval Expired, and a five minute
(:00-:05) when the RMF Interval Expired.
Change 37.136 ANALMSUS is a powerful set of reports of SOFTWARE MSUs
ANALMSUS from ASUMCELP, SMFINTRV, TYPE72GO, TYPE89 records, that
Jun 21, 2019 has many different bar charts, tabular reports, and EXCEL
spreadsheets as you choose, with report examples in the
comments, and with numerous report examples available
online at http://www.mxg.com/downloads/analmsus/
ANALMSUS.PDF
ASUMCELPMSU.XLSX
JOBSMSU.XLSX
MSU89.XLSX
REPORTCLASS.XLSX
SERVICECLASS.XLSX
TYPETASK.XLSX
(Don't be confused with archaic ASUMMSUS member.)
Change 37.135 Type 42 Subtype 5 Invalid LENSR values were individually
VMAC42 detected and LENSR=160 set, but now there are a total of
Jun 21, 2019 22 different values for records, because IBM populates
Jun 24, 2019 the total length and not the 160 first segment length.
But with 22 tests, I'm now forcing LENSR=160 always, as
that will ONLY fail if IBM actually changes that first
segment size in the future, you won't be ABENDing on
each new LENSR value. These are the known invalids:
IF LENSR IN (232,240,320,376,400,480,448,304,520,560,592,
720,640,1040,1120,1200,960,1360,1280,1440,880,800)
THEN LENSR=160;
Thanks to Robert Obee, Ensono, USA.
Change 37.134 CHART option changed to NONE because DSIG option is no
ANALACTM longer supported after SAS 9.3. The four coefficients
Jun 15, 2019 (CPU SRB MSO IOC) are added to WLM definitions report.
Change 37.133 TYPE42HI dataset, variables S42VSXST/S42VSXRT/S42VSXID
VMAC42 added in MXG 37.02 and MXG 37.03 incorrectly as character
Jun 17, 2019 variables with $EBCDIC8 informat. Change 37.019 in 37.04
corrected them to &PIB.4. numeric variables, but if you
build WEEKLY PDBs with some days created by 37.02/03 and
other days by 37.04 or later, you will need to either
DROP those variables from the 02/03 day's PDB, or just
remove the creation of TYPE42HI for that week.
It is always best if a new version of MXG is installed
to run on the first day of your week, so that all of
those daily PDBs will have identical structure.
Thanks to Jim S. Horne, Lowe's Companies, USA.
Change 37.132 Addition of a semi-colon following &OUTCODEx argument in
ASUM4HRS %VMXGSUM invocations to prevent rare errors.
GRAFCEC
GRAFWLM
VMXGRMFI
VMXGSUM
VMXGSUM
VMXGUOTT
Jun 14, 2019
Change 37.131 ANALCNCR fails with multiple errors if there are 0 OBS in
ANALCNCR the input datasets. Now detected and ANALCNCR ends.
Jun 21, 2019
Change 37.130 -New VM Account datasets supported.
EXVMCAPD dddddd Dataset Description
EXVMCAPE CHGD VMCAPCHD CAPABILITY*CHANGED
IMACVM CHGE VMCAPCHE CAPABILITY*CONTINUATION
TYPEVM -Under investigation with IBM Support:
VMXGINIT -Records with blank RECID (bytes 79-80).
Jun 25, 2019 -VMSESSN records overlap, with ACCTTIME greater than the
STARTIME=ACCTTIME-CONECTTM of the next record for the
User TCPIP, trying to use this data for availability
measurement.
Thanks to William Marshall, Ensono, USA.
Change 37.129 Member INSTALL and the listed members were updated with
INSTALL more consistent names in the examples.
MXGWPSV4 -The WPS V4 JCL Procedure need a new DD for 4.1:
JCLINSTW //MAPS DD DSN=&WPSHLQ..MAPS,DISP=SHR
MXGWPSV3
MXGWPSV4
JCLINSTL
JCLINSTT
CONFIGW4
Jun 12, 2019
Change 37.128 SMF 82 ICSF updates from ICN1633 for a future release:
FORMATS -New variable SMF82CSF in TYPE8201 identifies source of
VMAC82 the startup member name, formatted MG082CS.
Jun 12, 2019 -Variable SMF82UCB, SMF82TKF bits are decoded in TYPE8209.
-Variable SMF82TKF bits are decoded in TYPE8209
-Variable SMF82BOT bits are decoded in TYPE8213.
Change 37.127 -Formats created for TYP11902 dataset variables TTTTLSCS,
FORMATS TTTLSPD, TTTERMCD, TTSMCSTATUS, and values updated in
VMAC119 format $MG119RE for variable T119REAS.
Jun 11, 2019 -New TYP11902 variables:
TTSMCSTATUS='SMC-R*STATUS'
TTIPSECFLAGS='IP*SECURITY*STATUS'
TTLCLSMCBUFSZ='LOCAL RMB*BUFFER*SIZE KB'
TTRMTSMCBUFSZ='REMOTE RMB*BUFFER*SIZE KB'
-New TYP11994/TYP11995 OPENSSH new variables.
SSH_FIPSMODE ='RUNNING*IN*FIPS*MODE?'
SSH_KEXMETHOD ='KEY*EXCHANGE*METIOD*USED'
Thanks to Randy Hewitt, DXC, USA.
Change 37.126 z/VM MONWRITE deaccumulated field deltas are sometimes a
VMACVMXA negative value, especially in user fields like VMDTTIME,
Jun 7, 2019 usually related to a VM system event, but the original
assumption was that the negative value was due to a wrap
of the 4-byte accumulated value, so 4294967296 is added,
a guess at the full word wrap value, but these negatives
are not due to a wrap, and you get a very large value.
This change now sets the variable to a MISSING VALUE when
a negative delta is found, so those spikes won't impact.
Thanks to Terry Chao, DC Government, USA.
Change 37.125 Reserved Change.
====== CHANGES THRU 37.124 ARE IN MXG 37.04 DATED Jun 5, 2019 =========
Change 37.124 Variable ID added to the TYPE60,TYPE6156,TYPE62,TYPE64,
VMAC60 TYPE6367,TYPE68,TYPE69 datasets so a direct merge can be
Jun 5, 2019 made without added data passes.
Thanks to Tony Curry, BMC, USA.
Change 37.123 MXG calculation of TYPE70 variable CPUMVSTM/PCTMVSBY was
VMAC7072 too small because Parked Time was incorrectly subtracted
Jun 5, 2019 from CPUUPTM which already has Parked Time removed.
Variables PLCPRDYQ (Ready Queue Delay Percent) and
SHORTCPS were also too small and corrected. Impact was
typically less than ten percent.
Thanks to Ken Deering, COMPUWARE, USA.
Thanks to Selby Shanly, COMPUWARE, USA.
Change 37.122 Support for two new variables in RACF OFFLOAD RACF0200
VMACRACF dataset, with values of YES or NO:
Jun 3, 2019 USBD_ROAUDIT ='USER*HAS*ROAUDIT*ATTRIBUTE?'
USBD_MFA_FALLBACK='USE*PASSWORD*MFA UNAVAIL?'
Thanks to Karl Laseki, American Chemical Society, USA.
Change 37.121 Support for ThruPut Manager Release 18.02 v7r1.0.
VMACTPMX -New variables added to TPM10 dataset:
May 30, 2019 TPMCMLFL='TPMCMLFL*FLAG*BYTE'
TPMCMLCL='SLM*CAPACITY*LEVEL*1-5'
TPMCMLCP='CAPPED*PERCENT*LAST 5*MINUTES'
TPMCMLCC='CEC*CAPACITY*MSU/HR'
TPMCMLAG='AVG GS*JOBS*LAST*5 MIN'
TPMCMLAP='AVG PCS*JOBS*LAST*5 MIN'
TPMCMLAT='AVG GS+PCS*JOBS*LAST*5 MIN'
TPMCMSNM='LPAR*SET*NAME'
TPMCMSLM='LPAR*SET*LIMIT*MSU/HR'
TPMCMSA4='LPAR*SET*4HRAV*MSU/HR'
TPMCMSI5='LPAR*SET*5MINAV*MSU/HR'
TPMCMSFL='TPMCMSFL*FLAG*BYTE'
TPMCMSCL='LPAR*SET*CMP LIMIT*MSU/HR'
TPMCMSC4='CMP-WIDE*4HRAV*MSU/HR'
TPMCMSCI='CMP-WIDE*5MINAV*MSU/HR'
TPMCMSMA='MOBILE*4HRAV*MSU/HR'
TPMCMSBA='CATEGORY A*4HRAV*MSU/HR'
TPMCMSMI='CATEGORY B*4HRAV*MSU/HR'
TPMCMSMI='MOBILE*INTERVAL*USAGE*MSU/HR'
TPMCMSAI='CATEGORY A*INTERVAL*MSU/HR'
TPMCMSBI='CATEGORY B*INTERVAL*MSU/HR'
-New variable added to TPMSLM dataset
TPMSCLVL='MAXIMUM*CAPACITY*LEVEL'
Change 37.120 Mobile Service Units on GP and IIP ARE included in the
VMAC7072 CPUTM and ZIPCPUTM variables in TYPE72GO and TYPE72TR.
May 28, 2019 The comments in Change 36.253 are wrong and the proposed
CPUTM_ALL=SUM(CPUTM,CPUMOBILCP) is now CPUTM_ALL-CPUTM
and labeled EQUAL*TO*CPUTM. Using a WLM Policy that
classified the entire workload for a service class as
MOBILE, the Service Units were the same in the sum of
R723CCPU and R723CSRB (CPUTCBTM and CPUSRBTM), and in
R723TSUCP and in R723MSUCP (Total GP and Total Mobile).
Thanks to Ken Deering, Compuware, USA.
Thanks to Selby Shanly, Compuware, USA.
Change 37.119 Label for PTECP2 is 'CPU TIME*ZIP*ELIGIBLE' instead of
VMACNDM "QUALIFIED".
May 28, 2019
Thanks to Joe Faska, DTC, USA.
Change 37.118 Sites with NLS issues must use CONFIMXG, but to build the
JCLCONFI new FORMATS catalog, you must use the JCLCONFI example.
May 27, 2019
Change 37.117 The optional CICS DBCTL SMF 110 segment can be 164 or 256
IMACICDB but the order was the 164 first, so if you opened both of
May 23, 2019 the comment blocks, the 256 segment was misaligned. Now,
the 256 segment is first and both blocks can be opened to
support both lengths.
Thanks to Steven W. Erkkila, USBank, USA.
Change 37.116 -WPS U4087 ABEND in WPS 4.1 but not in WPS 4.0 due to the
CONFIGW4 new data copier added in 4.1, can be circumvented with
AUTOEXEW OPTIONS NOWPSSCATTERCOMP; which turns off the facility.
May 22, 2019 -This correction also fixed a CPU Loop in WPS 4.1.
Jun 17, 2019 -CONFIGW4 for z/OS and AUTOEXEW for ASCII have the option
added, but commented out, and for WPS 4.1 you must remove
the comment block. That option did not exist in 4.0.
Jul 3: Corrected in WPS 4.1.2.0.17535.
Change 37.116A Variable OPENTIME was repeated in _BTY1415 By List macro,
VMAC1415 causing NOTE:DUPLICATE BY VARIABLES. Second OPENTIME was
May 22, 2019 removed. Change 35.166 revised the BY list.
Thanks to Douglas C. Walter, CITIGROUP, USA.
Change 37.115 Wrong SMF record types for the example IFASMFDP step when
UTILBLDP BUILDJCL=YES was specified. 23 should have been 25 and
May 22, 2019 26J2/26J3 should be 26.
Change 37.114 Support for updated BETA 93 V6R2 (INCOMPATIBLE changes).
EXTYBET9 Offset to data was changed for some records.
EXTYBETP New subtypes create new datasets:
EXTYBETQ DDDDDD DATASET Description
EXTYBETR TYBET9 BETA9 RECORDS LIST/REPORT CVRTD
EXTYBETS TYBETP BETA12 PRINT HEADER PAGES
EXTYBETT TYBETQ BETA17 RECORDS MAILING OUTPUT
FORMATS TYBETR BETA30 DSC DATA CONVERTED LISTS
IMACBETA TYBETS BETA31 DSC RESRCS CVRTED LISTS
VMACBETA TYBETT BETA55 LOGOFF REQS WEB ENABLER
VMXGINIT Only BETA12 & BETA55 new datasets have been data tested.
May 22, 2019 Variable SYSUSRJOBCORR is INPUT and KEPT for subtypes
May 27, 2019 that contain it.
Thanks to Andreas Menne, Finanz Informatik-Sicherheitshinweis,GERMANY
Thanks to Martina Ruminski, Fin Informatik-Sicherheitshinweis,GERMANY
Change 37.113 NDM-CDI 24-byte record INPUT EXCEEDED ERROR; the header
VMACNDM length is 28 bytes, so a test for length is added and an
May 19, 2019 error message printed and the record deleted.
Thanks to Kurt Gramling, T-SYS, USA.
Change 37.112 New options CLEARALL=ONLY will clear any normal MXG
VMXGALOC associated LIBNAMEs allocated to your session without
May 19, 2019 trying to allocate new libnames.
Clears these libnames:
PDB SPIN MON TUE WED THU FRI SAT SUN WEEK
WEEK1-WEEK5 WTD MONTH MTD TREND
only if they are allocated.
Change 37.111 Final revisions for WSF/EOS WSFAUDIT variables AUDACT
FORMATS and AUDOBJN and their FORMATS, replaces Change 37.083.
VMACWSF -Dataset WSFACCT will always have zero observations; it
May 14, 2019 was never correct and is replaced by the four datasets
WSFDSN, WSFERD, WSFEVTSC, and WSFEVTPR.
Change 37.110 MXG Members TYPExxxx create output datasets in //WORK,
TECHNOTE MXG Members TYPSxxxx always SORT from WORK to PDB, and
May 11, 2019 the _Sxxxx sort macro all datasets for product xxxx and
deaccumulates those datasets with accumulated fields.
Exception: TYPEDB2 invokes the _SDB2 macro that sorts
all DB2 datasets except for DB2ACCT, and _SDB2
deaccumulates the DB2 datasets listed below
that need deaccumulation. They are also listed
in member DIFFDB2 lists sort/nonsort datasets.
NOTE: DB2ACCTP is sorted by _SDB2, but if you
only want Stats sorted, use _S100.
Exception: TYPE110/TYPE110S _S110 never sort these:
SUBTYPE=1, CICS MONITOR DATASETS:
_SCICTRN - CICSTRAN IS NOT SORTED, HIGH VOLUME
_SCICRDS - CICSRDS IS NOT SORTED, HIGH VOLUME
_SCICRDD - CICSRDPL IS NOT SORTED, HIGH VOLUME
_SCICRDF - CICSRDFI IS NOT SORTED, HIGH VOLUME
_SCICRDQ - CICSRDQU IS NOT SORTED, HIGH VOLUME
_SCICIDN - CICIDNTY IS NOT SORTED, HIGH VOLUME
_SCICIDD - CICIDNDD IS NOT SORTED, HIGH VOLUME
_SCICACC - CICSACCT NOT SORTED, PRE-CICS/ESA ONLY.
_SCICSYS - CICSYSTM NOT SORTED, PRE-CICS/ESA ONLY
PRODUCT DATASETS THAT ARE ACCUMULATED DDDDDD/DATASET
28 028IN7/NPMINPMT
30 TY30U6/TYPE30_6
50 DIF() ONLY FOR INTERVAL DELTA
79 TY791/TYPE791 TY792/TYPE792 TY799/TYPE799
TY79C/TYPE79C
99 TY99BG/TYPE99BG
102 102380/T102S380 102402/T102S402
103 TY1032/TYPE1032 TY103D/TYPE103D
108 TY1083/TYPE1083
110 INTTC/CICTC INTTSR/CICTSR INTDMG/CICDMG
INTVT/CICVT INTAUT/CICAUTO INTLDS/CICLDG
INTDTB/CICDTB INTTCR/CICTCR INTDQR/CICDQR
INTDQG/CICDQG INTTSQ/CICTSQ INTDS/CICDS
INTST/CICST INTFCR/CICFCR INTM/CICM
INTTDG/CICTDG INTSDG/CICSDG INTSMS/CICSMDSA
INTAUS/CICAUSS INTCO3/CICCONMR INTCO1/CICCONSR
INTDL3/CICDLIG INTDL1/CICDLIR INTDBU/CICDBUSS
INTPGG/CICPAUTO INTIRC/CICIRCB INTDMR/CICDMR
INTFEP/CICFEPIP INTFEC/CICREPIC INTFET/CICFEPIT
INTJCR/CICJCR INTLDR/CICLDR INTLS3/CICLSRFR
INTLS1/CICLSRR INTSDR/CICSDR INTSMD/CICSMD
INTSMT/CICSMT INTTC1/CICTCLR INTTDR/CICTDR
INTXMC/CICXMC INTUSG/CICUSG INTXMG/CICXMG
INTXMR/CICXMR
113 TY113/TYPE113 TY1131/TYPE1131
AIX ALL AIX Datasets
DB2 DB2PST/DB2PSTXX DB2NET/DB2NETXX DB2ST5/DB2STAT5
DB2ST0/DB2STAT0 DB2ST1/DB2STAT1 DB2STS/DB2STATS
DB2SBP/DB2STSBP DB2STB/DB2STATB DB2STR/DB2STATR
HSM HSMDSR/HSMDSRST HSMFST/HSMFSRTP HSMFUN/HSMDSRFU
HSMVSF/HSMVSRFU HSMVSR/HSMVSRST
IMS IMS452/4/6/7/8/9/C/D/E/F/O/P/G/H/I/J/K/L/M/N
UNS56B
MPLX MPLXIN/XSE/XGA/XRT/XPE/XPM/XPO
NDM NDMCT
ASI CPUTA_LF,TCBTA_LF,IOCNT_S,TRCA_S,TET,TRT
TCP TYTCPS/TYPETCPS
TPX TPXINT/TPXINTRV
VMXA SYTSYP/SYTPRP/SYTRSG/SYTRSP/SYTXSP/SYTASG/SYTSHS
SYTUSR/SYTCPC/SYTSCG/SYTCOM/SUTUWT/SYTSCP/SYTXSG
SYTCUG/SYTCUP/SYTCUM/SYTCPM/SYTSYG/SYTEPM/SYTLCK
SYTLCX/SCLADL/SCLDDL/SCLAEL/SCLSRM/SCLSTP/STORSG
STORSP/STOSHR/STOASP/STOBPG/STOXSG/STOXSU/STOASS
STOASI/STOSHD/STOVDK/USEDFC/USEATE/USEITE/PRCPRP
PRCIOP/PRCAPM/PRCMFC/PRCPUP/PRCMFM/IODDEV/IODMOF
IODVSW/VMDSES/ISFISA/ISFNOD/APLSRV/APLSLM/APLSLP
APLSL0/APLSLN/APLCMS/APLVMR/APLLXP/APLTC0/APLTC3
APLTC4/APLTC5/APLTC7/APLTC8/APLTC9/APLTCA/APLTCB
SSISCS/SSISMI/SSIXLK/SSIXDI
Change 37.109 Support for z/OS 2.4 SMF Manual 04MAR19 are already in
VMAC7072 place in MXG 37.02+, Change 37.037 from 14JAN19 Manual,
VMAC74 except for
May 10, 2019 -SMF70CPC_TYPE, listed in "Summary of Changes" page xxii,
but the field is not found in in the manual,query raised.
-New SMF70PRTCTV='SMF70OS*PRTCT*IS VALID?' flag in TYPE70.
compatibly added in this change.
Change 37.108 Bit mapping documentation for NDN-CDI CNF1/CNF2 fields:
VMACNDM NDMCNF1 $CHAR1. /*SECURE*COPY*FLAG1*/
May 9, 2019 /* BIT MAPPINGS FOR NDMCNF1 AND CISECNF1
PCEF EQU X'80' PNODE ENCRYPT.DATA
SCSI EQU X'40' SNODE SECURE.SIGNATURE
PCSI EQU X'20' PNODE SECURE.SIGNATURE
CCSI EQU X'10' COPY SECURE.SIGNATURE
SCEF EQU X'08' SNODE ENCRYPT.DATA
SSL EQU X'04' SSL.ENABLED=Y
TLS EQU X'02' TLS.ENABLED=Y
STS EQU X'01' STS.ENABLED=Y */
NDMCNF2 $CHAR1. /*SECURE*COPY*FLAG2*/
/* BIT MAPPINGS FOR NDMCNF2
CSIN EQU X'80' SIGNATURE = CURRENT KEY
PSIN EQU X'40' SIGNATURE = PREVIOUS KEY
TLS EQU X'20' TLSV10 ENABLED
STS EQU X'10' STS.ENABLED
IPV6 EQU X'08' IPV6 ADDRESS
TLS1 EQU X'04' TLSV11 ENABLED
TLS2 EQU X'02' TLSV12 ENABLED
ZFBA EQU X'01' ZFBA WAS USED */
NDMCPEA $CHAR1. /*MERGED*SECURE*ENCRYPT*NUMBER*/
Change 37.107 A change in the length of TPX05LEN misaligned TPXETIME &
VMACTPX TPXATIME; they incorrectly INPUT blanks, which TODSTAMP8
May 9, 2019 reported as 8am on Oct 27, 1935.
Thanks to Craig Bigler, Progressive, USA.
Thanks to Ann Knapik, Progressive, USA.
Change 37.106 A check of SYSFILRC that should have been inside a DO
BLDSMPDB loop checking SMFIN could cause a spurious critical error
May 8, 2019 saying that the allocation of the SMF file failed if some
May 13,2019 other earlier FILENAME statement had failed. FILENAME
May 19, 2019 statements don't tell us when they have a problem until
you try to use them unless you check the SYSFILRC macro
variable for a non-zero value.
-If you run a weekly job independently of a daily job and
are using AUTOALOC=YES and need to rerun the week using
FORCEDAY it pointed at the incorrect day and did not
recognize the start of the week. FORCEDAY should always
be the date of the data being processed so if your week
starts on Monday FORCEDAY should point at Sundays date.
If you are running a weekly or monthly job the code
validating parameters still looks at the value in
BUILDPDB and if it did not match what was expected could
cause a failure. Now you can either omit the parameter
and allow it to default or you can specify BUILDPDB=NO.
-If you run TREND daily and needed to rerun a WEEK, the
trending ran as if it were daily. If you are using
AUTOALOC this just repeats what was already done and
there will be no duplication of data. Now BLDSMPDB checks
to see that RUNDAY is NE NO.
-BLDSMPDB now sets SYSCC=16 if it detects any errors, and
displays that condition code value at the end.
Change 37.105 Support for SMF 120 WAS and LIBERTY COMPATIBLE new data:
VMAC120 -Subtype 11. TYP120BL. SM120BDL='ON IF*CVTZCBP*IS ON?'
May 6, 2019 -Subtype 09. TYP1209N. SM1209HW='ON IF*CVTZCBP*IS ON?'
SM1209HX='WORKER*THREADS*PRESENT'
-Subtype 12. TYP12012. SM120CEJ='ON IF*CVTZCBP*IS ON?'
Change 37.104 Variables CECSER and CPCMODEL are added to TYPE72GO data
VMAC7072 set, retained from prior 70. However, they are set blank
May 6, 2019 if the PREVVSYS system is not the SYSTEM of this record,
May 19, 2019 which could happen if the SMF data was sorted before MXG
or if an SMF Dump happens to start with type 72 records.
Thanks to Andrew Petersen, DXC Technology, AUSTRALIA.
Change 37.103 Support for IMS Log Records 5607/5610/5904/5950 creates
EXIMS567 new datasets:
EXIMS56A DDDDDD DATASET DESCRIPTION
EXIMS594 IMS567 IMS5607 MCS/PICOS
EXIMS595 IMS56A IMS5610 START PHASE 1 SYNCPOINT
IMACIMS IMS569 IMS5609 CCTL DISCONNECT FROM DBCTL
VMACIMS IMS56B IMS5611 END OF PHASE 1 SYNCPOINT
VMXGINIT IMS56F IMS5615 RRS RESTART DONE
May 15, 2019 IMS594 IMS5904 REGION OCCUPANCY RECORD
IMS595 IMS5950 DATA BASE LOG RECORD
Change 37.103A FORMAT $MGFSMFID updated for DB2 102 IFCIDS for ANALID.
ANALID
FORMATS
May 15, 2019
Change 37.102 Support for CICS/TS 5.5 new Statistics, COMPATIBLE, two
EXCICMPR new datasets, and all _SCICxxx sorts now deaccumulate.
EXCICSJN For 5.5, fields were inserted into reserved areas.
FORMATS -New Dataset CICMPR for STID=145 CICS Policy statistics.
IMAC110 MPR_POLICY_NAME ='POLICY*RESOURCE*NAME'
VMAC110 MPR_RULE_NAME ='POLICY*RULE*NAME'
VMXGINIT MPR_POLICY_USERTAG='POLICY*USERTAG'
May 6, 2019 MPR_BUNDLE_NAME ='POLICY*BUNDLE*NAME'
May 18, 2019 MPR_BUNDLE_DIR ='POLICY*BUNDLE*DIR'
Jun 2, 2019 MPR_RULE_TYPE ='RULE*TYPE'
MPR_RULE_SUBTYPE ='RULE*SUB*TYPE'
MPR_ACTION_TYPE ='ACTION*TYPE'
MPR_ACTION_COUNT ='RULE*ACTION*COUNT'
MPR_ACTION_TIME ='RULE*LAST*ACTION*TIME'
-New Dataset CICSJN for STID=150 NODEJSAPP statistics.
SJN_NAME ='NODEJSAPP*NAME'
SJN_LE_RUNOPTS ='NODEJSAPP*LE*RUNOPTS'
SJN_STATE ='NODEJSAPP STAT'/
SJN_DEFINE_SOURCE ='GROUP*INSTALLED*FROM'
SJN_CHANGE_TIME ='CHANGE*CREATE*TIME'
SJN_CHANGE_USERID ='CHANGE*USERID'
SJN_CHANGE_AGENT ='CHANGE*AGENT'
SJN_INSTALL_AGENT ='INSTALL*AGENT'
SJN_INSTALL_TIME ='INSTALL*CREATE*TIME'
SJN_INSTALL_USERID ='INSTALL*USERID'
SJN_CREATION_LCL ='CREATION*TIME*LOCAL'
SJN_PID ='NODEJSAPP*PID'
SJN_BUNDLE_NAME ='BUNDLE*NAME'
SJN_CPU ='TOTAL*CPU*TIME'
SJN_HEAP_CURRENT ='ALLOCATED*HEAP'
SJN_HEAP_RUNTIME ='HEAP*USED BY*RUNTIME'
SJN_HEAP_APP_DATA ='HEAP*USED FOR*DATA'
SJN_HEAP_MAX ='MAX*POSSIBLE*HEAP'
SJN_INVK ='COMPLETED*INVOKES'
SJN_INVK_ERR ='COMPLETED*INVOKES*IN ERROR'
SJN_INVK_CUR ='CURRENT*INVOKES*IN PROGRESS'
SJN_INVK_PEAK ='PEAK*INVOKES IN*PROGRESS'
SJN_NODEHOME ='NODEHOME*PROFILE*ENTRY'
SJN_PROFILE ='PROFILE'
SJN_STARTSCRIT ='ENTRY*JAVASCRIPT'
SJN_STDERR ='STDERR*FILE'
SJN_STDOUT ='STDOUT*FILE'
SJN_TRACE ='TRACE*FILE'
SJN_LOG ='LOG*FILE'
-Dataset CICDB2GL STID=102 new variable
D2GTCBPR='TCB*PROTECTED*CURRENT'
-Dataset CICCONSR STID=52 new variables.
A14EAHWM='MAX*AIDS'
A14EALL expanded to 4 bytes, used reserved area.
-Dataset CICMNG STID=81 new variables
MNGIR ='IDENTITY*RECORDS'
MNGIRS ='IDENTITY*RECORDS*SUPP BY EXIT'
MNGDPLRL='DPL*RESOURCE*LIMIT'
MNGURIRL='URIMAP*RESOURCE*LIMIT'
MNGWEBRL='WEBSVC*RESOURCE*LIMIT'
-Dataset CICXMR STID=11 variable
XMRAENDC='ABEND*COUNT'
-Previously _SCICddd Statistic Dataset Sort Macros only
PROC SORTed from WORK to PDB; there was no deaccumulation
so fields with accumulated values were wrong. Now, all
_SCICddd macros de-accumulate correctly into the PDB.
The _S110 macro sorts account and all statistics datasets
the _S110ST macro sorts only the statistics datasets.
-By DEFAULT, TYPE110 & BUILDPDB do NOT invoke _S110ST. All
datasets are left in work, where you can tailor EXPDBOUT
to sort all or individual datasets. You can use
%LET EPDBOUT= _S110ST ; in your SYSIN to sort the stats
datasets AND deaccumulate to correct errors in CICINTRV.
-But if UTILBLDP is used to create your tailored BUILDPDB,
and if CICS data was requested, then _S110 is invoked, so
your PDB.CICINTRV will be valid as soon as you use 37.04.
-TYPS110 invokes _S110, TYPE110 does not.
-Revised deaccumulation logic needed JOB READTIME added
to the BY list, and logic NOT FIRST.READTIME used to
eliminate large values created when back-to-back regions
had forward times.
-A new _SCICxxx sort macros option MXGCICRQTSORT can be
used to only read and use the SMFSTRQT='INT' interval
records with this statement in your //SYSIN:
%LET MXGCICRQTSORT=%QUOTE(WHERE SMFSTRQT='INT');
This needs testing when you have multiple RQTs.
The MXG default continues to use ALL record types.
-A new macro variable &MXGCICSORTED is set to YES in
_S110 and _S110ST macros so that the logic in VMXGCICI
knows to use the PDB deaccumulated data. You would only
need to set it to YES if you are building CICINTRV in
a separate job from the one that created the stats PDB.
Change 37.101 MXG 37.03, if you used USERADD=102.nnn syntax for DB2
UTILBLDP IFCID subtype, and used BUILDPDB=YES, the generated code
May 6, 2019 was wrong and failed with 455-185 W102nnn error.
Thanks to Tim Hare, Florida Department of Transportation, USA.
Change 37.100 DB2 zPARM T102S106 vars were wrong in V11/V12 because
VMAC102 QWP4CYR input $EBCDIC8 but it is only one byte:
May 3, 2019 QWP4CYFR QWP4DDLM QWP4CDSTL QWP4ZHYPL QWP4STACS
These zPARM variables in DB2 V12 are now supported:
QWP4RTNP ='REORG*TS_NOPAD*DEFAULT?'
QWP4DYNPFSW ='QWP4DYNPFSW'
QWP4PSPN ='PAGESET*PAGENUM*ABSOLUTE*RELATIVE?'
QWP4RDS_DM_BLKFI='QWP4RDS*DM*BLKFI'
QWP4NIDX ='QWP4NIDX'
QWP4IXMC ='INDEX_MEMORY_CONTROL'
QWP4UHMDH ='UTILS*HSM_MSGDS*HLQ'
QWP4DINA ='DEFAULT*INSERT*ALGORITHM'
QWP4MISD ='QWP4MISD'
QWP4FLT ='QWP4FLT'
QWP4IXMT ='QWP4IXMT'
QWP4AUTC ='AUTH*COMPATIBILITY'
QWP4TSCT ='QWP4TSCT'
QWP4ENKL_OFF ='OFFSET*FOR*ENCRYPTION*KEYLABEL'
QWP4CDRL='COMPRESS*DIRLOB'
QWP4SFPR='STATFDBK*PROFILE'
QWP4AUTCSU='SELECT*FOR*UNLOAD'
Thanks to Lai Fai Wong, Bank of America, USA.
Change 37.099 Two formats for CICS Version variable SMFPSRVR displayed
FORMATS 72 instead of 'TS5.5' or '5655-Y04 in MOBILE data.
May 2, 2019
Thanks to Scott Barry, SBBWorks Inc., USA.
Change 37.098 -Possible Abend S0C7 when using the CDF Filter SYSPLEX=
ASMRMFV after ASMRMFV change 36.241. Affects MXG releases
May 1, 2019 36.12-37.03. HAS NOT OCCURRED, exposure was observed.
-Options message RMFV037I incorrectly shows SHOWASI,
but SHOWASI option is not actually in effect. Affects
MXG release 37.03 only.
Change 37.097 APAR OA56762 NEGATIVE SMF30_TIME_ZIIP_ON_CP zOS 2.2 only,
TECHNOTE FLASH caused LARGE CPU time of 42,949,672 seconds because MXG
Apr 30, 2019 input as PIB4 expecting positive values. The INVALID
DATA BIT in SMF30TF2 for this time field WAS NOT ON.
This is variable CPUZIETM='ZIP-ELIGIBLE*CPU TIME*ON CP'
in MXG TYPE30 datasets, and the defect was in eight
subtype 3 interval termination records in this SMF file
Thanks to Jutta Gleixner-Schmid, ALLIANZ. GERMANY.
Change 37.096 RMF III dataset ZRBASI variables ASIFRXB_LF,ASIFRXA_LF
VMACRMFV and ASIFRXH_LF are the sum variables that should have
Apr 26. 2019 been divided by ASISMPCT to report their average value.
The labels are also corrected. Variables CPC_CECNAME
and LPARNAME are added to dataset ZRBBDSIH.
Thanks to Karl Laseki, American Chemical Society, USA.
Change 37.095 New variables added to TYPEDBDS (IMF from BMC):
FORMATS -DBTRIOTM DBTWIOTM DBTFLAG2 DBTFLAG3
VMACCIMS DBTNOI DBTNOO DBTBFSTK DBTBFSTN
VMACIMS -Formats for DBTFLAG2 and DBTFLAG3 created.
Apr 25, 2019 -IMS07 ENDTIME could be missing due to 8 bytes
May 3, 2019 found but not documented; detection/protection
was added.
-Variable BHTOON is added to CIMSTRAN and CIMSDBDS;
it was already kept in CIMSPROG.
-Variable ALPCPTRN was incorrectly formatted $HEX8 and
incorrectly used to create UOWTIME.
-UOWTIME was incorrectly created like CICS UOWTIME with
only 6-bytes of datetime, but IMS UOWTIME is 8-bytes in
UOWTRANS, now used to create the IMS UOWTIME.
Thanks to Randy Hewitt, DXC, USA.
Change 37.094 CICS 110 Stats CICLSRR dataset accumulated variables that
VMAC110 end with BFF/CRF/CRS/CWF/CWS/FRD/UIW are now correctly
Apr 24. 2019 deaccumulated in _SCICLS1 sort macro when TYPS110 is used
Apr 26, 2019 or it can be added after TYPE110 is included.
See Change 37.102, all CICS Statistics are deaccumulated.
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.093 Support for IMS ODBM Accounting SMF Type 29 Subtype 1
EXTY29OD record creates the new TY29ODBM dataset.
VMAC29
Apr 24, 2019
Thanks to Kurt Gramling, T-SYS, USA.
Change 37.092 TRENDINCODE= parameter added to both macros to allow you
VMXGRMFI to limit the trend datasets. You could limit the amount
VMXGDBSS of data retained by specifying (using VMXGDBSS here) by:
Apr 20, 2019 TRENDINCODE=IF DATEPART(BEGTIME) GE TODAY()-732;
to impose a 2 year limit on the data.
====== CHANGES THRU 37.091 ARE IN MXG 37.03 DATED Apr 19, 2019 =========
Change 37.091 -TYPE42 Subtype 5 INPUT EXCEEDED when new MCO/SYO/BGO data
VMAC42 segments exist, MXG miscoded the new segments, MXG 37.02.
VGETUTKN -Hex 00 protection for UTKNPOE/UTKNSCL/SMF42GAO/SMFA2GAO/
Apr 18, 2019 SMF42FAJ/SMF42FBN/SMFA2FAJ/SMFA2FBN/SMF42GAN/SMFA2GAN.
ST 5 Error caught by Early Adopter tester in MXG 37.03EA.
-Jun 17: Variables S42VSXST and S42VSXRT were created in
MXG 37.02-03 incorrectly with $EBCDIC8 informat. This
change corrected them to &PIB.4. numeric variables in
TYPE42HI dataset, but if you build WEEKLY PDBs with some
days 37.02/37.03, and some 37.04 or later, you will need
to DROP those variables from the 02/03-built PDB.
Change 37.090 When BUILDPDB=YES is used with %UTILBLDP, the SMF 113
UTILBLDP records are automatically processed, and ASUM113 is run
Apr 15, 2019 after BUILDPDB. Now, SUPPRESS=113 can be used if you
don't want those datasets.
Change 37.089 -Major enhancements to ASMRMFV and VMACRMFV.
ADOCRMFV -New Field Data Filter (FDF) feature supports field level
ASMRMFV data selection for the RMF III ASI, CSR, and DVT tables.
VMACRMFV -The ANDIF=, ORIF=, IF= keywords are used to specify a
Apr 16, 2019 FDF filter called an IF expression.
-FDF complements the existing Character Data Filter (CDF)
feature. CDF has no numeric data filtering capability.
-If both CDF and FDF filters are used, then CDF filtering
occurs first. Entries filtered by CDF are never passed
to FDF.
-FDF supports character, fixed point, floating point,
percent, bit string, hex string, time, and time of day
fields for filtering depending on the format and content
of each field.
-FDF uses IF expressions with 3 components enclosed in
required matched left/right parentheses in this order:
1. RMF III field name or MXG variable name when
supported
2. Comparison operator
3. Character, numeric, bit string, hex string, percent,
time, or time of day value as appropriate to the
field being filtered.
-See documentation Sections 31-34 for full details on the
use of FDF.
-New RMFV080I-RMFV088I messages added for FDF support.
-New options SHOWARR (alias SHARR) and NOSHOWARR (alias
NSHARR) added to display IF expression memory storage
array activity. Default is NOSHOWARR. see Section 6
Report Control Parameters for more details.
-New options SHOWIF (alias SHIF) and NOSHOWIF (alias
NOSHIF) added to display IF expression compare results
in both Detail and Summary reports in message RMFV080I.
Default is SHOWIF. see Section 6 Report Control
Parameters for more details.
-New option IFERR= controls handling of errors detected
while processing ANDIF=/IF=/ORIF= expressions. Possible
settings are ABEND, ERROR, WARN, IGNORE. Default is
IFERR=ERROR. See Section 8 Error Handling Parameters
For more details.
-Three recently documented RMF Monitor III tables are now
supported for PDB builds: PCI, SCM, and ZFS.
-The PCIE Activity Data Table may be selected with the
PCI, P, MOST, or ALL table select options.
-The Storage Class Memory Data Table may be selected with
the SCM, MOST, or ALL table select options.
-The ZFS Performance Data Table may be selected with the
ZFS, Z, MOST, or ALL table select options.
RMF III data set.
-ASMRMFV will now detect quoted strings and not apply any
translations regardless of UPCASE/NOUPCASE settings.
-ASMRMFV will no longer check for Control Unit Busy or
Switch Port Busy when filtering DVT entries with the
NOZEROIO option. These DVT fields became obsolete and
unused with z/OS 1.4 in September 2002.
-Almost all documentation for ASMRMFV now resides only
in the ADOCRMFV member.
-ADOCRMFV has been reformatted to take advantage of full
72 column width for better legibility.
-Message RMFV033* showed an incorrect value for index
count if an I/O error occurs reading the first or last
MINTIME interval.
-Many documentation sections have been updated and 4 new
sections are added:
Section 31 Field Data Filtering (FDF)
Section 32 Filtering The ASI Table
Section 33 Filtering The CSR Table
Section 34 Filtering The DVT Table
Change 37.088 DB2 102 IFCID 106 truncated variables longer lengths are
VMAC102 now supported:
Apr 15, 2019 QWP4SADM='INSTALLATION*SYSTEM*ADMIN*USERID'
QWP4DFID='SYSTEM*DEFAULT*USERID'
QWP4ADM2='SYSTEM*ADMIN*ID2'
QWP4OPR1='MVS*OPERATOR*ID'
QWP4OPR2='MVS*OPERATOR*ID2'
QWP4REGC='DDL*REG*TABLE*OWNER'
QWP4REGA='DDL*REG*ART*NAME'
QWP4REGO='DDL*REG*ORT*NAME'
QWP4OZUS='ONLINE*ZPARM*USERID*MONITOR'
QWP4FCCD='UTILS*FCCOPYDDN*PARM*DEFAULT'
Thanks to Lai Fai Wong, Bank of America, USA.
Change 37.087 SMF 50 VTAM Tuning record subtypes 02 and 05 don't match
VMAC50 the record documentation and IBM has acknowledged and
Apr 15, 2019 will revise their doc, when this text will be updated.
This change only reverses the order of RDUX/REDE.
Thanks to Svend Zaunick, Finanz Informatik, GERMANY.
Change 37.086 MXG Support for the new Japanese Reiwa era dates is in
TECHNOTE place as MXG does not use any Japanese informats, but
Apr 10, 2019 SAS Note 63973 reports an update is needed to provide
support in NENGO and JNENGO informats/formats.
Change 37.085 SMF 92 Subtype 52 INPUT EXCEEDED. because SMF92TRSN was
VMAC92 documented as 52 EBCDIC on page 846, but there are only
Apr 10, 2019 8 bytes at the end of the record for the name, and the
segment length is 48 to match a final 8-byte field.
But I believed the SMF manual and INPUT 52 without an
extra test to see if the bytes were there.
Thanks to Joe Faska, DTCC, USA.
Change 37.084 Variables CMB10C0-CMB10C4 in VXPRCAPM are wrong because
VMACVMXA their DIF() calls had the wrong variable.
Apr 10, 2019
Thanks to Graham Harris, RBS, ENGLAND.
Change 37.083 -Ignore the first two sections of this original text.
VMACWSF -WSF/EOS revisions corrected misalignment in WSFAUDIT
VMACWSF dataset, but AUDOBJT values of '6C'x and 'B4'x are not
Apr 10, 2019 documented, and the order of AUDACT and AUDOBJT is NOT
Apr 19, 2019 consistent with the documentation, which has always has
May 14, 2019 ACT then OBJT, but for ACT values of '60'x or higher,
I've observed OBJT is first, and the value of OBJT has
to be used to decode the multiple uses of the AUDOBJI
field into the correct variable.
-Also, OBJT values of '6C'x,'B4'x are found but not
documented in the DSECT.
-Dataset WSFACCT will always have zero observations; it
was never correct and is replaced by the four datasets
WSFDSN, WSFERD, WSFEVTSC, and WSFEVTPR.
-May 14, 2019. See Change 37.111.
Change 37.082 SMF 99 ST 12 dataset TYPE99CC Capacity Increase/Decrease
VMAC99 bit variables S99CCCCAPINCR/S99CCCCAPDECR/S99_VCM_D2-4
Apr 9, 2019 are decoded into individual variables:
S99CCCCAPINCR0='ADJUST*CAPACITY*INCREASE?'
S99CCCCAPINCR1='ADJUST*CAPACITY*INCREASE*BY UNPARK'
S99CCCCAPINCR2='UNPARK*REQUEST?'
S99CCCCAPINCR3='UNPARK*ALL*REQUEST?'
S99CCCCAPINCR4='RESERVED'
S99CCCCAPINCR5='UNPARK*CAPACITY*BELOW?'
S99CCCCAPINCR6='CAPPED*UNPARK*HIGH VH*UTILIZATION?'
S99CCCCAPINCR7='RESERVED'
S99CCCCAPDEC00='ADJUST*CAPACITY*DECREASE?'
S99CCCCAPDEC01='ADJUST*CAPACITY*DECREASE*BY UNPARK'
S99CCCCAPDEC02='PARK*REQUEST?'
S99CCCCAPDEC03='PARK*ALL*REQUEST?'
S99CCCCAPDEC04='MVSBUSY*TOO*LOW?'
S99CCCCAPDEC05='VL*EFFECT*TOO LOW?'
S99CCCCAPDEC06='SMALL*VM/VL*EFFECTIVENESS?'
S99CCCCAPDEC07='NO*VM/VL*EFFECTIVENESS?'
S99CCCCAPDEC08='IF*NO*VH*EXISTS?'
S99CCCCAPDEC09='NO DECREASE*LOW*CEC*UTILIZATION?'
S99CCCCAPDEC10='PR/SM*CAPPED*PARK ALL?'
S99CCCCAPDEC11='PR/SM*CAPPED*PARK ALL*HI CEC UTIL?'
S99CCCCAPDEC12='PR/SM*CAPPED*VH UTIL*LOW?'
S99CCCCAPDEC13='PR/SM*CAPPED*VL*EFFECT*TOO LOW?'
S99CCCCAPDEC14='PR/SM*CAPPED*MVS BUSY*TOO LOW?'
S99CCCCAPDEC15='PR/SM*CAPPED*ADJUST*CAPACITY*DECR?'
S99CCCCAPDEC16='PARK ALL*REQUEST*unpark*threshold?'
S99CCCCAPDEC17='PR/SM CAPPED*NO DECR*LOW CEC UTIL?'
Thanks to Jan Tielemans, KBC, BELGIUM
Change 37.081 Velocity VPS USER records are either Interval or Summary
VMACXAM but only the top ten users get Interval records, so when
Apr 5, 2019 analyzing the USER data,you MUST select IF INTORSUM='SU'
to see the total resource usage.
Thanks to Deeresh Naidoo, First Rand Bank of South Africa.
Change 37.080A Datasets BETA9706 and BETA9706D were not output to PDB
VMACBE97 when TYPSBE97 was used to sort from work to PDB.
Apr 4, 2019
Thanks to Andreas Menne, Finanz Informatik, GERMANY.
Change 37.080 Variables SVPCNM Service Class and RPRTCLAS Report class
VMACRMFV are added to all the RCD datasets.
Apr 4, 2019
Thanks to Claudio A. Rodriguez, BancoGFalicia, ARGENTINA
Change 37.079 New variables DBS_DD and DBS_D are created in TYPETPMX.
VMACTPMX Variables SYSTEM,SMFTIME added to ERROR messages.
Apr 2, 2019 Variable $JCL_S decoded and output, Apr 19.
Apr 19, 2019
Thanks to Jack Hyde, OPTUM, USA.
Change 37.078 BY variable R748SIID in dataset TYPE748 format is now
VMAC74 HEX4 (was HEX2), and there are no duplicate observations
Apr 2, 2019 in TYPE748S as R748SIID is unique to each record, due
to that too-short format, false duplicates could have
been deleted in the past.
Thanks to Douglas C. Walter, CITRIBANK, USA.
Change 37.077 Enhanced to sort and remove intervals where a SYSTEM is
SAGANAL on multiple CECs, as when it was moved from one CEC to
Apr 3, 2019 another. New time range report of input SMF 30/70s.
Thanks to Bob Berg, American Family, USA.
Change 37.076 Support for HSM FSR Record addition of Unix filename in
VMACHSM dataset HSMFSRBO variable FSR2_UNAM because FSRDSN is
Apr 2, 2019 only 44 bytes, When FSR2_UNML is greater than 44, FDRDSN
will contain the first part of the name, then ... and
then the last part of the name, while FSR2_UNAM will
contain the full name (up to 128 char).
Thanks to Harald Seifert, Huk-Coburg, GERMANY.
Change 37.075 INPUT STATEMENT EXCEEDED and INVALID DATA for SM125THA
VMAC125 because INFORMAT &PIB.2. was missing the final period.
Mar 30, 2019
Thanks to MP Welch, Bank of America, USA.
Change 37.074 QBSTBPIN was always incorrectly calculated before the
VMACDB2 variables used in the calculation had been DIF()'d
Mar 30, 2019 yielding unrealistically high values. The calculation is
now done in DIFFDB2 after the DIF() calls are done.
Thanks to Randy Hewitt, DXC, USA.
Change 37.073 BBMQ processing reported UNEXPECTED RTINs, when you have
VMACBBMQ multiple BMC products writing to a common history file.
Mar 29, 2019 MXG BBMQ now selects only E1x-E8x, skipping other values
eliminating those log messages. There were no changes
to BBMQ 5.3 records in 5.4.
Thanks to James Wajda, Credit-Suisse, USDA.
Change 37.072 ODS Statistical graphics procedures make extensive use
TECHNOTE of JAVA, which can be very memory intensive on zOS. This
Mar 28, 2019 is any procedure starting with SG or any ODS HTML or ODS
PDF outputs. SAS recommends REGION=512M minimum but the
MXG recommendation is always REGION=0M on the JOB card
on zOS. We have seen REGION usage as high as 900M+ for
jobs running the ANAL9914 Topology report with HTML.
Change 37.071 %MACRO variables INTIME70,INTIME70EN,INTIME70PR can be
VMXG70PR used by ASUM70PR to convert time zones of LPARs to a
VMXGINIT common timezone. This example shifts all times to GMT.
Mar 27, 2019
%LET INTIME70=
%QUOTE(
STARTIME=STARTIME-GMTOFFTM;
SMF70GIE=SMF70GIE-GMTOFFTM;
MACHTIME=MACHTIME-GMTOFFTM;
);
%LET INTIME70EN=
%QUOTE(
STARTIME=STARTIME-GMTOFFTM;
SMF70GIE=SMF70GIE-GMTOFFTM;
);
%LET INTIME70PR=
%QUOTE(
STARTIME=STARTIME-GMTOFFTM;
SMF70GIE=SMF70GIE-GMTOFFTM;
MACHTIME=MACHTIME-GMTOFFTM;
);
%INCLUDE SOURCLIB(ASUM70PR);
This is an initial design, which may be revised.
Thanks to Berthold Willing, AXA, GERMANY.
Change 37.070 Unused Change Number.
Mar 25, 2019
Change 37.069 zEDC Compression types for DCOLLECT datasets revised:
FORMATS Dataset DCOLDSET, variable DCDCTYPE formated values
VMACDCOL 0=0:Not Compressed
Mar 25, 2019 1=1:Generic
2=2:Tailored
3=3:ZEDC
DCDCTYPE replaced DCOLMTYP, now always missing.
Dataset DCOLDC, variable DDCCT format MGDCOCT:
0=0:Generic
1=1:Tailored
2=2:ZEDC
Change 37.068 CF Activity Report Structure Level is moved to ANALRMF3
ANALRMF3 from VMACRMFV. Additional RMF III report examples will
Mar 21, 2019 be added in ANALRMF3.
Change 37.067 Support for RMF III PCI, SCM, ZFX segments create four
EXZRBPCI datasets ZRBPCI, ZRBSCL, ZRBZFX (System Data) and
EXZRBSCL ZRBZFS (File Server Data). ZRBSCL was used for SCM
EXZRBZFX because there already is a ZRBSCM dataset (for CFISSCMS).
EXZRBZFS
IMACRMFV
VMACRMFV
VMXGINIT
Mar 29, 2019
Change 37.066 New TOKEN variables added to TYPE80TK dataset:
VMAC80A TOKMCTOKENKY TOKMCTOKENTM TOKMSISCCNO TOKMEMPLID
Mar 21, 2019
Thanks to Mark Kerr-Delworth, Express ICS, ENGLAND.
Change 37.065 Major rewrite of this macro to eliminate repeating the
VMXGDSN same logic 3 different times. New parameter added to ID
Mar 19, 2019 your HSM managed tape volumes with a default of HSM.
Labels on VOLUMES and TAPE variables corrected to show
that they are actually counts of datasets and not a
count of volumes.
Change 37.064 Doc ONLY. Examples add to suppress 110.1 or 101 or both
UTILBLDP when BUILDPDB=YES.
Mar 20, 2019
Change 37.063 -The ANALID report showed only IDANDSUM=26.000 for either
ANALID JES2 or JES3; now the SUBS (2 for JES2, 5 for JES3) is
VMACID stored in SUBTYPE to create 26.002:JES2 or 26.005:JES3.
Mar 16, 2019 -A non-impacting note about DELETE SMFREC01/02 removed.
Change 37.062 Faulty logic prevented creation of zip eligible chart.
GRAFWRKX
Mar 15, 2019
Change 37.061 Bar charts of ZIP and ZIP eligible added.
GRAFWLM
Mar 15, 2019
Change 37.060 SMF Type 82 subtype 31 INPUT STATEMENT EXCEEDED for
VMAC82 TAG='0204' because MXG incorrectly expected 8 bytes when
FORMATS that TAG only has 8 bytes. FORMAT MG082SN revised to
Mar 20, 2019 include SERVER name.
Thanks to Randy Springs, BB&T, USA.
Change 37.059 Further corrections for IFCID 319 support.
VMAC102
Mar 10, 2019
Change 37.058 Cosmetic: Uninitialized variable CBPERROR.
VMAC30
Mar 11, 2019
====== CHANGES THRU 37.057 ARE IN MXG 37.02 DATED Mar 11, 2019 ========
Change 37.057 Analysis of what your WLM Classification Rules do, using
ANALRULE SMF 30, 101 and 110 to produce three reports for where
Mar 10, 2019 work is sent by SYSTEM and TYPETASK, routing for CICS
transactions by SYSTEM and APPLID, and routing for DDF.
Change 37.056 Unused Change.
Change 37.055 RMF III dataset ZRBCFI, CFACT Coupling Facility Activity
FORMATS Report, which is actually a CF per-Structure report, is
VMACRMFV printed by invoking _CFACT after TYPSRMFV. Macro _CFACT
Mar 9, 2019 is defined at the bottom of VMACRMFV.
Mar 21: MOVED TO NEW ANALRMF3 Report Member. CH 37.068.
Thanks to Ervin Claxon, CSX, USA.
Change 37.054 Using report classes to define workloads in RMFINTRV is a
UTILWORK good way to group workloads but will only work if all
Mar 9, 2019 workloads have a default report class, UTILWORK now
detects this condition and warns that the use of report
classes should not be attempted until this can be
resolved
Change 37.053 New utility contribution, UTILMISS, will create a data
UTILMISS set from an existing dataset, removing all variables that
Mar 8, 2019 have all numeric missing values, and characters blank.
Thanks to Jim S. Horne, Lowe's Companies, USA.
Change 37.052 Variable CPUERROR in TYPE30 datasets is a two-byte field
VMAC30 but was accidentally made length $4 some time ago due to
Mar 7, 2019 a blank that gave it $HEX8 format which forced length 4.
The correct $HEX4. format is now applied, but LENGTH $4
is also forced, since a change in length will cause PROC
APPEND to fail if the user didn't specify FORCE. These
TYPE30xx datasets are too critical to not protect.
-Flag variables SMF30T32 and SMF30T33 are now kept.
Change 37.051 -IFCID 319 variable QW0319FL is replaced by a variable
VMAC102 for each bit:
Mar 7, 2019 QW0319UR='CALLER*PASSED*USER*REG*NAME?'
QW0319AE='AES*ENCRYPTION*BEING*USED?'
QW0319SC='COMPATIBLE*WITH*TCPALVER?'
QW0319SE='SECURE*CONNECTION?'
-Variable QW0319RI is INPUT and Kept.
-Variables QW0319AE QW0319IY QW0319SC QW0319SE kept.
-New variable QW0319LU='LUNAME*IF*SNA' is created
Thanks to Warren Cravey, FMR, USA.
Change 37.050 UTILBLDP now accepts USERADD=100 or 101 or both, and
UTILBLDP invokes USERADD=DB2, but if only 100 or 101 alone are
Mar 6, 2019 requested, the other record's datasets are _NULL_ed.
UTILBLDP now accepts USERADD=CICS (to read SMF 110)
Change 37.049 Variable ASIQSCANREQ is kept in RMF III ZRBASI dataset.
VMACRMFV
Mar 6, 2019
Change 37.048 Example to "BUILDPDB" only JOB-related Datasets enhanced
JCLPDBJB to support both JES2 and JES3.
Mar 5, 2019
Change 37.047 NDM-CDI dataset NDMCT variable NDMCPU was 256 times too
VMACNDM large and character NDMRIUP6/NDMTYPFK shifted because of
Mar 1, 2019 a 1 byte misalignment in the MXG Input Statement.
Thanks to Mike Creech, BKFS, USA.
Thanks to Roger Foreman, BKFS, USA.
Thanks to Glenn Halligan, BKFS, USA.
Thanks to David Kelley, BKFS, USA.
Change 37.046 Variables SMF70BPS/SMF70ACS for each SMF70CIN engine
VMAC7072 type are variables CP70BPS/IFA70BPS/ZIP70BPS/IFL70BPS
Mar 5, 2019 and CP70ACS/IFA70ACS/ZIP70ACS/IFL70ACS in PDB.TYPE70PR.
and PDB.ASUMCELP (recommended LPAR analysis) and ASUM70LP
(which is BY SYSTEM and has duplicate data.
Thanks to Scott Barry, SBBWorks Inc., USA.
Change 37.045 The CTG Version, CTGRVN was added to each of the CTG
VMAC111 datasets created from SMF 111 Records and variables
Mar 5, 2019 CTGIALRQ CTGLCNFA are kept in dataset TY111CXI.
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.044 The BMC CMF Product generally updates VERSNRMF only on a
VMAC7072 CMF Release Boundary; values of both 792 (2.2) and 794
VMAC71 (2.3) exist on z/OS 2.3. MXG does NOT use VERSNRMF for
VMAC73 any logic, but this correction for CMF records sets the
VMAC74 VERSNRMF=794 if it was 792 and RMFSRCL Record Level is
VMAC75 81 or 82 depending on subtype.
VMAC76
VMAC77
VMAC78
VMAC79
Mar 1, 2019
Thanks to Joe Faska, DTCC, USA.
Change 37.043 Executing MXG on ASCII under a VM product, or with WORK
TECHNOTE on a network drive, bad things can happen. We SRONGLY
Feb 21, 2019 recommend keeping the WORK file local to the system on
which you are executing SAS.
Two Known Errors:
-This error was found with WORK on a network drive:
ERROR: Unable to obtain valid utility file pathname.
-This error was found under VM with WORK on a network
drive:
ERROR: A lock is not available for WORK.OPTVAR.DATA.
This one can be circumvented by adding
-filelockwait 30
to your SAS command or as an OPTION in SAS.CFG file.
Change 37.042 Change 37.024 did not protect for a blank WANTONLY, but
READDB2 only generated a cosmetic error that %SYSFUNC did not
Feb 19,2019 have right number of arguments.
Change 37.041 DCOLLECT APAR OA54879 reports that DCDEXFLG is NOT USED
VMACDCOL for zEDC compression and is now "DATA SIZES NOT VALID'
Feb 19,2019 and is for non-VSAM Extended Format Data Sets.
Data set sizes that are not valid in either or both of
DCDUDSIZ or DCDCUDSZ variables, which might contain non
zero values, but should not be used.
Thanks to Robert Chavez, FPL, USA.
Change 37.040 Dataset TYPE749 variable R7491DEFCOMPRATIO wrong value
VMAC74 is corrected to
Feb 19,2019 IF R7491IOB GT 0 THEN
R7491INFCOMPRATIO=R7491IIB/R7491IOB;
ELSE R749INFCOMPRATIO=.;
Variables were reversed in the calculation.
Thanks to Jan Tielemans, KBC, BELGIUM.
Change 37.039 Three MQ reports matching IBM's MQSMF program. Currently
ANALMQ the reports are a queue summary, a detail report of PUTS,
Feb 14,2019 and a detail report of GETS. The GETS/PUTS reports can
be output to either or both SYSOUT or CSV files.
Thanks to Jan Tielemans, KBC, BELGIUM.
Change 37.038 Simple report that combines catalog records (TYPE6156)
ANALHSM1 with HSM activity (HSMFSRST) to generate a report of
Feb 14, 2019 datasets that may be thrashing between primary and
migration.
Thanks to Richard Way, Office Depot, USA.
Change 37.037 Updates found in the Jan 14, 2019 SMF Manual,
FORMATS with more changes to come in 37.03 when validated.
VMAC124 -New Datasets
VMAC41 TYPE42VO='PER-VOLUME STATISTICS'
VMAC42 TYPE42HI='HIGH RESPONSE TIME JOBS'
VMAC7072 -TYPE70 New variables
VMAC74 SMF70_IPL_TIME ='IPL*DATETIME*OF*PARTITION'
VMAC78 SMF70_TRG_M_COUNT='TRG*MEMORY*CONSUMPTION*SAMPLES'
Feb 19, 2019 -TYPE70TR New variable
TRG_MEM ='TENANT*MEMORY*CONSUMPTION*MGBYTES'
-TYPE7002 New variables
R7023SCOPE='80X=CPC SCOPE*40X=SYSTEM*SCOPE'
R7023DID='DOMAIN*ID'
-TYPE70X2 New variables
R7024SCOPE='80X=CPC SCOPE*40X=SYSTEM*SCOPE'
R7024DID='DOMAIN*ID'
-TYPE70Y3 New variables
R7025SCOPE='80X-CPC SCOPE*40X=SYSTEM*SCOPE'
R7025DID ='DOMAIN*ID'
-TYPE749 New Variables
R749LKID='SYNC*I/O*LINK*IDENTIFER'
-TYPE78IO Change
R783DSTX relocated.
-TYPE41 Change
SKIP LOGIC protection for future, no impact now.
-TYPE 42 SUBTYPE 5 New Variables
S42VRID1='DELAYS*TIME*1-10 MICROSEC'
S42VRID3='DELAYS*TIME*100-10000 MICROSEC'
S42VRID4='DELAYS*TIME*1-10 MILLISEC'
S42VRID5='DELAYS*TIME*10-100 MILLISEC'
S42VRID6='DELAYS*TIME*OVER*100 MILLISEC'
S42VRIDM='MAXIMUM*I/O*INTERRUPT*DELAY TIME'
S42VRIDT='DATETIME*OF THE*MAXIMUM'
S42VRIDA='AVERAGE*I/O*INTERRUPT*DELAY TIME'
S42VRBSY='TOTAL*BUSY*TIME'
S42VRRSP='COMMANDS*DELAYED*BASE*RESERVED'
S42VRRSN='CHANNEL*PROGRAMS*WITH*RESERVE'
S42VRRES='DURATION*WHEN*RESERVED'
S42VRREX='LONGEST*CONTINUOUS*RESERVED'
S42VRRSR='AVERAGE*RESPONSE*PROGS WITH*RESERVE'
-TYPE 42 SUBTYPE 6 New Variables
S42SNAVGARDELAY ='AVG APPLICATION*RESUME*DELAY'
S42SNARDELAYCOUNT='AVG APPLICATION*RESUME*DELAYS'
S42DXMXI ='STORAGE*SUBSYSID*FOR S42DSMXR'
-TYPE124 New variable
SM124S1WWPN='WORLDWIDE*PORT*NAME'
Change 37.036 Some IFA variables were not populated in the four output
VMAC7072 datasets created by ASUM70PR; all IFA variable names are
VMXG70PR unchanged, but all "ZAAP" text in labels is now "ZCBP".
Mar 5, 2019
Change 37.035 DB2 V12 overlooked Package Variables now INPUT and KEEP:
VMACDB2 QPACAWLH ='LATCH*WAIT*TIME'
Feb 7, 2019 QPACANLH ='WAITS*TRACE*EVENTS*LATCH'
Mar 5, 2019 QPACRLNU ='THREADS*TO ROLL DATA'
QPACAACC ='WAITS*TRACE*EVENTS*ACCELERATOR'
QPACAACW ='WAIT TIME*ACCELERATOR*REQUESTS'
QPAC_PQS_WAIT ='WAIT TIME*TO SYNC*PARALLEL*QUER'
QPAC_PQS_COUNT ='SUSPENDS*WAITING*SYNC*PARALLEL'
QPAC_PIPE_WAIT ='WAIT TIME*PIPE'
QPAC_PIPE_COUNT ='WAITS*FOR*PIPE'
QPAC_COPYID ='PACKAGE*COPY*ID'
-Macros _N100/N101/_S100/_S101 defined for UTILBLDP.
Change 37.034 Two more SMF 42 subtype 5 ABENDING invalid LENSR=520/592
VMAC42 length values added to the test. APAR OA54668 corrects.
Feb 7, 2019 LENSR IN(232,240,320,376,400,480,448,304,520,560,592,640)
Change 37.033 Support for Axway V3.3.2 2018/06/27 restructured User
VMACAXWY SMF record.
Feb 6, 2019
Thanks to Warren Cravey, FMR, USA.
Change 37.032 TYPE749 z/EDC data, DIVIDE BY ZERO protection failed if
VMAC74 both R7491DCT and R7491ICT were zero; R741BPS had been
Feb 5, 2019 included in the test, but it's always non-zero so the
test now is IF SUM(R7491DCT,R7491ICT) GT 0 THEN
R749FPGBPRT=(1048576*R7491BPC/((R7491DCT+R7491ICT)*R7491BPS));
Thanks to Harald Seifert, Huk-Coburg, GERMANY.
====== CHANGES THRU 37.031 WERE IN MXG 37.01 DATED Feb 3, 2019=========
Change 37.031 Bar charts of MIPS and % CPU added to the analysis work
GRAFWLM by IMPORTANCE, originally based on Peter Enrico's paper.
Feb 3, 2019
Change 37.030 MXG 36.12-37.01 ASMRMFV ASI NOZEROCPU filter stopped
ASMRMFV filtering which could cause a significant increase in the
Feb 3, 2019 size of the RMFBSAM file and the PDB.ZRBASI dataset.
ASICPUTA_LF was added by Change 36.241 to that filter,
but it is an accumulated field that can not be used as it
is always non-zero. NOZEROCPU is supported for z/OS 2.2+
-Section 5 "Input Data Control Parameters" is updated.
====== CHANGES THRU 37.029 WERE IN MXG 37.01 DATED Feb 1, 2019=========
Change 37.029 A reference line of the values for SMF70LAC (overall
ANAL89 rolling 4 hour avg MSU) added to interval MSU charts.
Feb 1, 2019
Change 37.028 Support for z/VM 7.1 (INCOMPAT, BROKEN CONTROL RECORD)
VMACVMXA due to insert in VXPRCDHF plus the change in HCPCPEID
Jan 31, 2019 value for the Service Level test from '40061802 for 6.4
Feb 14, 2019 to '10071802' for 7.1 that failed when tested for 'GE'.
New variable ZVMVERS='07.1.18.1' is created so GE can be
used for IF ZVMVERS GE tests. There were 32 bytes added
to PRCDHF, but the 7.1 DSECT only shows on byte added.
-Heuristic (ZZQUCT+1) test revised, false positive caused
large VMDTTIME value.
Thanks to Graham Harris, RBS, ENGLAND.
Change 37.027 Example added to email the final condition code of a SAS
EMAIL job running in the background on ASCII.
Jan 31, 2019
Change 37.026 If you create your own SUBSYSTEM that handles JES2 output
VMAC26J2 the SUBS value in SMF 26 records is not the expected 0002
Jan 30, 2019 for JES2 (or 0003 for JES3), but instead is a two byte
character field. Previously, MXG TYPE26J2 only output
SUBS=2 execution purge records. Now, if the SUBS is 3,
the JES3 record is deleted with MXGERROR messages that
TYPE26J3/BUILDPD3 must be used for JES3. If the SUBS is
other than 2, the records are presumed to be valid JES2
records, but MXGWARN messages print the SUBSYSTEM name.
Thanks to Randy Hewitt, DXC, USA.
Change 37.025 CICS variable A11ACTCI in CICS Statistics dataset CICDQG
VMAC110 was INPUT but not KEPT.
Jan 29, 2019
Thanks to Giuseppe Giacomodonato, EPVTECH, ITALY.
Change 37.024 Using READDB2 with WANTONLY arguments DB2ST/DB2PST caused
READDB2 ERROR: Char Operand in the %EVAL, due to a superfluous
Jan 29, 2019 AND in an %IF statement that also exposed that PDBOUT=YES
Feb 1, 2019 failed in the PROC SORT of the DB2STATx datasets.
The WANTONLY doc has been updated to note that
WANTONLY DOES NOT SUPPORT DB2STATS DB2STATR DB2GBPST
DB2STATB DB2SDTSBP AND DB2STAT5 DATASETS.
Thanks to Keith C. Shaffer, Cigna, USA.
Thanks to James Cyr. Cigna, USA.
Change 37.023 Unused Change Number.
Change 37.022 The new $MGRMIPS format that maps IBM processor model to
GRAFWRKX MSU and MIPS capacity (created in 37.001) is now used in
GRAFCEC these Graphs, which use PDB.ASUMCELP which has variable
Jan 28, 2019 CPCFNAME so MIPS can be calculated from percent busy:
MIPS=PCTBUSY/100*PUT(CPCFNAME,$MGRMIPS.);
The CPCFNAME is constructed from CPU TYPE and MODEL as
CPCFNAME=PUT(CPUTYPE,$HEX4.)!!'-'!!CPCMODEL;
programmatically, or manually as '3906-420' for that z14.
Thanks to Ervin Claxon, CSX, USA.
Change 37.021 If you run on ASCII with autoaloc=yes and did not put an
VMXGALOC execution of VMXGALOC in your IMACINIT (strongly our
Jan 28, 2019 recommendation) but added a second execution of VMXGALOC
and the parameters did not precisely match those you used
in BLDSMPDB and did not specify READONLY=YES to suppress
the aging off of old directories you could potentially
lose data as it deleted old directories. This change now
looks to see it you specified READONLY as NO and the
current days PDB has already been created it will
generate an MXGNOTE and set READONLY to YES.
-YR2KEEP and BASEYEAR parameters are added to let you
allocate a yearly PDB, Defaults to 0, so it won't be
created or allocated unless you enable it, and you will
need to update your BLDSMPDB to add the dataset names
you want created in the yearly directories. Contact
support@mxg.com if you need help.
Change 37.020 Unused Change Number.
Change 37.019 Documentation note for VMXGUOW/ASUMUOW. If you have a
VMXGUOW tailored CICSTRAN dataset and have dropped any variables,
Jan 25, 2019 you may get an UNITIALIZED message for those variables in
ASUMUOW. While this is an expected and non-critical error
you can remove the error by editing the _KUOWIDC _KUOWCIC
or _Kuowcix macros where the variables appear and simply
remove the variable(s) from the list.
Change 37.018 Support for STC HSC Subtype 32 & 33 create new datasets
EXSTCV32 STCVSM32='RESERVED,INTERNAL USE'
EXSTCV33 STCVSM33='MVCPOOL USAGE'
VMXGINIT
VMACSTC
Jan 27, 2019
Thanks to Randy Hewitt, DXC, USA.
Change 37.017 Many non-fatal corrections were made to type 92:
VMAC92 -Several 16-byte STCKE datetimes inputs were wrong:
Jan 24, 2019 SMF92CCT,SMF92MCT,SMF92CCT,SMF92FSMN
-GMT92OFF had to be relocated around the STCKE INPUT.
-Subtype 50+ have 72 byte data section, 32 for LT 50.
-Subtype 50 variables LRP-LRN only in SMF92EVENT=1,
and INPUT changed from &PIB.4.3 to &PIB.4.0.
-Subtype 50 variables OVS/OCH only in SMF92EVENT=4
-Poor Labels SMF92EVENT/SMF92VOL/SMF92CCHH/SMF92VCN
were enrichened.
-Subtype 50 SMF92OCH corrected format to HEX for CCHH.
-Variable SMF92ADN only kept in TYPE9217 dataset.
-Subtype 59 only the first instance was output.
-Subtype 52 and 54 had incorrect SKIP values.
-Subtype 58 SMF92TRL relocated outside the loop.
-Subtype 59 2 byte filler removed with STCKE fix.
This demonstrates the difficulty in writing new code
with no SMF records to test. SMF 92 code was updated
in Aug 2017 in Change 35.180, expecting a user to send
SMF data if problems were observed, and that didn't
happen until January 2019, with most of the above fixes
mostly done by these two users:
Thanks to John Compton, World Programming Ltd, ENGLAND.
Thanks to Steve Bagshaw, ITMetrics, ENGLAND.
Change 37.016 Report showing total bytes/counts and min/max datetimes
ANALID for each SYSTEM is added as the second report.
Jan 19, 2019
Change 37.015 Variable EDGRTIME had missing values with DATEFORM=E/A/I;
FORMATS RHDTFORM logic moved ahead of EDGRTIME for H record and
VMACEDGR label corrected to EDGRTIME='REPORT*DATE TIME'
Jan 20, 2019
Thanks to Lindsay Oxenham, Department of Defence, AUSTRALIA.
Change 37.014 INPUT EXCEEDED DB2STAT1 SMF 100 Subtype 1 NETEZZA/IDAA
VMACDB2 record because IBM changed the length of OFFQ8ST segment
Jan 18, 2019 but couldn't change LENQ8ST because it is a single field
in the header that should apply to all segments. But the
correct length is now set with LENQ8ST=Q8STNAME_OFF+8;
using the end of the name field for the actual length.
Thanks to Graham Harris, RBS, ENGLAND.
Thanks to Randy Hewitt, DXC, USA.
Change 37.013 New example to count both TAPEDRVS and STEPS, and the
ANALCNCR Concurrency with only one pass of the data.
Jan 20, 2019
Thanks to Randy Hewitt, DXC, USA.
Change 37.012 z/VM Service Level 40061802 INPUT STATEMENT EXCEEDED due
VMACVMXA to VMMTRSYS inserting 60 bytes in the 1.04 record.
Jan 17, 2019
Thanks to Craig S. Bigler, Progressive Insurance, USA.
Change 37.011 Variables QXFETCH/QWACSPEB UNINITIALIZED due to misspell.
VMXGUOW Enhanced to make it easy to only process CICSTRAN data:
Jan 19, 2019 IF _LDB2ACC=_NULL_, DB2 data will not be read.
Jan 25, 2019 IF _INMQ=_NULL_ MQ data will not be read.
Counts of OBS before and after are created and if the
OBS reduction is LT 2, an MXGNOTE advises you to skip
using ASUMUOW which can be very CPU intensive and is
really needed for heavy CICS MRO sites, to consolidate
those multiple CICSTRAN observations into one UOW,
Unit of Work observation.
Thanks to Gary Keeres, Indianapolis Power & Light, USA.
Change 37.010 Reserved Change Number.
FORMATS
VMAC89
Jan 16, 2019
Change 37.009 The $%VGETOBS(DDNAME=&PDBMXG should be (DDNAME=&PDBMXG1
GRAFCEC although no error occurred unless you had set a value
Jan 14, 2019 other than "PDB" for the location of the input PDB.
Thanks to Tom MacCabe, Dominion Energy, USA.
Change 37.008 Enhancement, addition of INCODE= parameter to allow
GRAFWRKX selection by date or system with your inserted code.
Jan 11, 2019 Suppressed a no longer needed graphics catalog note.
Change 37.007 -Support for Beta93 Version 6.2 subtypes 2 and 3, which
VMACBETA both have a lot of undocumented data: subtype 2 docs 140
Jan 22, 2019 but length is 208 and subtype 3 docs 156 with 224 length.
Change 37.006 -If MXGDEBUG has length GT 0 and DSNSTRING or DATASET are
VMXGPRAL zero, debugging messages are created by VMXGPRAL.
VMAC102 -Unmatched parens in data set labels read by VMXGPRAL
Jan 11, 2019 caused non-fatal error messages for datasets with obs:
ERROR: Expected close parenthesis after macro function.
All MXG dataset's labels were examined and VMAC102 for
IFCIDs 84 85 86 87100 101 174 AND 175 were corrected.
Change 37.005 Support for DB2 102 Trace IFCID/SUBTYPE 404 populates
VMAC102 T102S404 (Authorization Compatibility) dataset with
Jan 10, 2019 QW0404xx variables.
Thanks to Warren Cravey, FMR, USA.
Change 37.004 Reading z/OS DATA using the SAS FTP Access method needs
TECHNOTE the RCMD='SITE RDW' argument:
Jan 9, 2019 FILENAME SMF FTP ("'SYS1.SMF'" "'SYS2.SMF'" ... )
USER='XXXXXX' HOST='YYYYYYY' DEBUG
S370VS RCMD='SITE RDW' LRECL=32760
PASS='XXXXXXXX' PASSIVE;
If RCMD is not used, the transfer will time out when
PASSIVE is specified, or will produce a RACF ERROR
if PASSIVE was not specified.
Change 37.003 SMF119 dataset TYP11952 variable SMF119ML_IP_IPV4 was
VMAC119 wrongly compressed of blanks from variable TIRIP instead
Jan 8, 2019 of from SMF119ML_IP_IPV4 in line 4519.
Thanks to Randy Shumate, RELX Group, USA
Change 37.002 -SMF92 Subtype 8 INPUT STATEMENT EXCEEDED because the
VMAC92 documented length of SMF92GDD in the SMF Manual is 4,
Jan 8, 2019 but the length in the record is 8 bytes.
-There were also numerous non-fatal corrections:
-SMF92 Subtype 50, SMF92STHCL missing period, and names
SMF90OIOCCL/SMF82VCX were corrected to SMF92.
Thanks to Miroslav Kubik, IBM Corporation, CZECH REPUBLIC.
Thanks to John Compton, World Programming Ltd, ENGLAND.
Change 37.001 -Improve ability to propagate variables of interest to
ADOCRMFV multiple SAS data sets during RMF III PDB build.
ASMRMFV -RMF III CPCDB table fields CPC_CecMSU, CPC_LparMSU, and
FORMATS CPC_HomeLPName, are added to MXG01 record from the first
VMACRMFV MINTIME interval for each RMF III VSAM dataset at open,
Jan 6, 2019 but those MXG01 values are the values only at start.
Jan 14, 2019 -Format $MGRMIPS is created to map GEIMODEL/GEIIMDL to
Jan 29, 2019 create the CPC_CECMIPS using IBM's LSPRITR table:
Feb 2, 2019 IBM Resource Link: Large Systems Performance Reference
https://www-01.ibm.com/servers/resourcelink/
lib03060.nsf/pages/lspritrzOsv2r2?OpenDocument
-RMF III GEI table fields GEIMODEL and GEIMDL added to
MXG01 record from the first MINTIME interval for each
RMF III VSAM data set, but the MXG01 value is only the
RMF III start. But by RETAINing these fields from the
GEI and CPU records, and by relocating the write of the
GEI, CPU, and CPCDB records before the ASI records for
each interval, the CEC MODEL/Speed and MSU/MIPS capacity
variables CPC_CECNAME CPUCEMSU CPUCEMIPS CPULPMSU
GEIMODEL GEIIMDL and LPARNAME are populated in ZRBASI
and several other datasets.
-NOTE: Not all the fields mentioned above are always
static. Use the RMF III DATASET SWITCH option to force
RMF III to overwrite the oldest data set whenever it is
started if separation of configuration changes by data
set is desired.
-CSVQUERY macro added to show load module size in Change
36.241 for ASMRMFV destroyed pointer to SCT (Step Control
Table). This caused ASMPGM and ASMSTEPN fields in MXG00
record to be garbage and also in message in VMACRMFV.
-Section 12 "Messages" is updated.
-In MXG 36.36 the fields GEIIMDL/GEIMODE were added to
ZRBASI, but they could be incorrect, precipitating
this change in logic in VMACRMFV.
LASTCHANGE: Version 37.