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

EXQAPSTY QAPMFRLY Frame Relay Data

IMACQAPM QAPMSTND DDI Station Counter Data

VMACQAPM QAPMSTNY Frame Relay Station

Nov 20, 1993 Changes were compatibly made, so prior versions of MXG

will not fail with the version's data records.

Change 11.253 Support for Memorex Telex LMS Version 2.17 SMF record now

EXLMSCUP creates these two new datasets:

EXLMSDAL Dataset Subtype Description

IMACLMS LMSCUP 06 Cart Update (inhibit SCR indicator)

VMACLMS LMSDALC 29 Device Allocation

Nov 20, 1993 Also, VMACLMS was restructured for consistent colimation.

Thanks to Dan Kaberon, Hewitt Associates, USA
Change 11.252 Support for Mobius INFOPAC-RDS user SMF record is added

ADOCIPAC by this comprehensive user contribution. Datasets are

ANALIPAC documented in ADOCIPAC, and sample analysis reports are

EXIPAC01 in ANALIPAC! Five new datasets are created:

EXIPAC02 Dataset Subtype Description

EXIPAC03 IPAC01 01 Batch Printing Usage Statistics

EXIPAC04 IPAC02 02 Online Printing Usage Statistics

EXIPAC05 IPAC03 03 Online Viewing Usage Statistics

IMACIPAC IPAC04 04 Archive Recall Usage Statistics

TYPEIPAC IPAC05 05 Archive I/O Subsystem Tuning & Perf

VMACIPAC And the analysis example, ANALIPAC, is comprehensive: it

Nov 19, 1993 combines the VERSIONS and DISTRIBUTION databases from

the INFOPAC-RDS product, with MXG's PDB.HSMFSRST (from

HSM), and PDB.PRINT (from JES2 printing), Vanguard's

RACF database, with three of the five SMF subtypes.

Thanks to Jeff Burnett, Anixter, USA.

Change 11.251 The ASM program to break down MVS/ESA 4.3 RMF Monitor III

ASMRMFV segments into actual records was missing code (lost in

Nov 19, 1993 fax transmission and transcription!). The code has been

corrected, and renamed to ASMRMF3 instead of ASMMON3.

This is preliminary for MXG support of the compressed ESA

4.3 data - the SAS code to process the output of the ASM

utility is still in development. See Change 11.238.

The ASMRMF3 member was replaced by ASMRMFV.

Change 11.250 Xerox SFS records from 9700 INVALID ARGUMENT TO INPUT

VMACSFS FUNCTION occurs if CUSTJNR contains hex nulls instead of

Nov 18, 1993 blanks. Insert this line:

VAR(10)=TRANSLATE(VARS(10),'40'x,'00'x);

immediately preceding the existing line:

CUSTJNR=INPUT(VARS(10),8.);

Thanks to Doug Medland, Confederation Life Insurance, CANADA.
Change 11.249 Support for Integris' UniKix product, used to downsize

EXTYUNIA CICS VSAM and DB2 applications to the Unix environment.

EXTYUNIK The product creates a binary file of performance records

IMACUNIA and provides a conversion program to create a portable

IMACUNIK ASCII file. MXG can process either the binary or ASCII

TYPEUNIA file under UNIX, Windows, or OS/2 SAS, or the ASCII data

TYPEUNIK can be uploaded for processing under MVS SAS. This is

VMACUNIA the beta release of the UniKix measurements, and Integris

VMACUNIK will restructure the data records in their next release

Nov 17, 1993 (at my suggestion) so they will be true variable-length,

self-defining records so that future enhancements can be

made to the data records compatibly, so the MXG support

will be revised when those changes are available.

The current implementation members ....UNIA are for the

ASCII file (infile name UNIKIXAS), and members ....UNIK

are for the Binary file (infile name UNIKIXBI).

Change 11.248 Support for BatchPipes/MVS type 91 SMF record creates 11

EXTY9101 new datasets:

EXTY9102 Dataset Subtype Description

EXTY9103 TYPE9101 01 Subsystem Initialization

EXTY9104 TYPE9102 02 Subsystem Interval

EXTY9111 TYPE9103 03 Subsystem Termination

EXTY9112 TYPE9104 04 Subsystem Parms Changed

EXTY9113 TYPE9111 11 Pipe Connection Open

EXTY9114 TYPE9112 12 Pipe Connection Interval

EXTY9115 TYPE9113 13 Pipe Connection Close

EXTY91IC TYPE9114 14 Pipe Create

EXTY91OC TYPE9115 15 Pipe Delete

IMAC91 TYPE91IC 11-13,15 Input Connection Details

TYPE91 TYPE91OC 11-13,15 Output Connection Details

VMAC91 Subtypes 1 and 3 are Pipes Subsystem event records.

Nov 17, 1993 Subtypes 2 and 4 are interval or termination records with

statistics on each Pipes Subsystem (pipes/connections

created/deleted/active, input/output bytes, etc.).

Subtype 11-15 are Pipe Connection records, which identify

the pipename being accessed, and (for subtype 12 and 15)

contain pipe activity counts. In addition, subtypes 11,

12, 13, and 15 can contain one or more Connection Detail

sections (either Input or Output connection), so MXG

creates one observation in TYPE91IC or TYPE91OC for each

Connection Detail segment (which contains JOB,READTIME,

DDNAME, six timestamps, reads/writes, bytes in/out, etc.)

for each user of a pipe. This structure lets you analyze

overall pipe statistics with the subtype datasets, or the

pipe statistics by user with the IC/OC detail datasets.

MVS Pipes looks to be a very powerful new MVS facility!

Change 11.247 Support for Novell Network Navigator user SMF record now

EXTYNNAV creates one dataset:

IMACNNAV Dataset Description

TYPENNAV TYPENNAV Network Navigator Session Resources

VMACNNAV with session logon and logoff datetimes, JESNR, bytes

Nov 16, 1993 sent/received, and TGETS/TPUTS for the session.

Thanks to Elena Beryozkin, United Parcel Service, USA.
Change 11.246 Support for NetView Performance Monitor NPM Version 2.1.0

EX028NWC added major enhancements, especially in the area of VTAM

EX028NWD monitoring, with CPU and storage used by VTAM. Thirteen

EX028RMA new datasets are created:

EX028RMD Dataset Subtype Description

EX028VAP NPMRMSTR BA RTM ACTIVATE

EX028VAD NPMRMSTP BB RTM DEACTIVATE

EX028VBF NPMVTSTR D0,D1,D2 VTAM START/STOP/MD

EX028VEN NPMVTEXC D3,D4 VTAM EXCEPTION

EX028VDV NPMVSVEN D5 VTAM ENVIRONMENTAL

EX028VGB NPMVSVGB D6 VTAM GLOBAL

EX028VTE NPMVSVBF D7 VTAM BUFFER POOL

EX028VTS NPMVSVDV D8 VTAM DEVICE DATA

EX028VVR NPMVSVVR D9 VTAM VIRTUAL ROUTE

FORMATS NPMVSVAP DA VTAM APPLICATION

IMAC28 NPMVSVAD DB VTAM ASID DATA

VMAC28 NPMNWCWC FC WRKSTN-NPM CONNECT

Nov 13, 1993 NPMNWDWD FD WRKSTN-NPM DISCON

Change 11.245 The original change described in this text in MXG 11.07

VMXGSUM thru MXG 11.09 were never actually implemented. The text

Nov 12, 1993 of the change was written, but the VMXGSUM member was not

moved into the MXG library, because the KEEPIN= logic did

not produce identical output datasets, and it had to be

backed out of the changes. See Change 11.281 & 11.309.

Change 11.244 A new utility, UCICSCNT counts how many SMF type 110 CICS

UCICSCNT records come from each CICS Region, by CICS version, and

Nov 10, 1993 by record subtype, so you can prove to your CICS guru

which APPLIDs have monitoring and/or statistics records.

Thanks to Richard S. Ralston, Whirlpool Corporation, USA.
Change 11.243 Support for NETWISE's RPC EXEC product type 33 SMF record

VMAC33 adds these "accounting" fields to dataset TYPE33_1:

Nov 9, 1993 NETWCAPN,NETWCOMP,NETWCORR,NETWMDNM,NETWRPCM,NETWRPCN,

NETWRPCN,NETWRPCT and NETWSAPN. MXG now creates one obs

in TYPE33_1 for each TPUSE section; previously only the

first TPUSE section's data was output. With NETWISE's

data, there are two sections in each record, one for the

inbound and one for the outbound connections, so there

two observations for each NETWISE SMF record.

Thanks to Linda Marzik, EDS Performance and Evaluation Group, USA.

Change 11.242 DB2 variable NETSNAME still can have blanks, after Change

VMACDB2H 11.050, preventing DB2ACCT matching with CICSTRAN. This

Nov 8, 1993 happens if the last half of NETSNAME contains blanks; the

correction is to translate all blanks in DB2 NETSNAME to

nulls. After the NETSNAME algorithm, and just before

"/* CREATE UOWTIME FROM QWHCTOKN */", insert the line:

NETSNAME=TRANSLATE(NETSNAME,'00'X,'40'X);

Thanks to Edward McCarthy, Health Insurance Commission, AUSTRALIA.

Change 11.241 TRNDRMFI will fail with VARIABLE PLATCPUS NOT FOUND when

TRNDRMFI ASUM70PR is not executed. Add in the INCODE= section:

Nov 8, 1993 IF PLATCPUS=. THEN PLATCPUS=.; just like PARTNCPU.

There is no requirement to run ASUM70PR before TRNDRMFI,

but without it, PLATCPUS did not exist, until the change.

Thanks to Tom Elbert, John Alden Life Insurance, USA.

Change 11.240 Variables READY12 READY13 READY14 READY15 were left out

TRND70 of the NORM2= list in this trending member, causing their

Nov 8, 1993 summary values to be erroneously large.

Thanks to Tom Elbert, John Alden Life Insurance, USA.

Change 11.239 MXG Early PreRelease 11.08 only. IF ID=80 SOURCE='RACF';

ANALSMF should have been IF ID=80 THEN SOURCE='RACF';

Nov 8, 1993

Thanks to Freddie Arie, Lone Star Gas, TEXAS.

==Changes thru 11.238 were included MXG Early PreRelease 11.08 01Nov93=
Change 11.238 This new Assembler program reads the RMF Monitor III VSAM

ASMRMFV files from MVS/ESA 4.3 (which are compressed, and are not

Oct 30, 1993 standard records). It was typed from a fax, and has not

been tested, but I wanted to make it available for tests

because there are reported problems in the ZRB processing

that are still unresolved. See Change 11.251.

Was ASMMON3 until change 11.251.

Thanks to Don Friesen, B.C. Systems, CANADA.

Change 11.237 ANALDB2R could not use a PDB on tape for reports, because

ANALDB2R multiple datasets were opened simultaneously, but with a

Oct 30, 1993 new TREND= parameter, either the Accounting or Statistics

reports can be run from your PDB= on tape. Only if you

want to include your TREND data base and your PDB data

bases together for a report will you need to point TREND=

to the DDname for your TRNDDB2A, TRNDDB20, and TRNDDB21

datasets.

Change 11.236 Internal logic was improved for performance. If the first

VMXGSUM data step or the SORT is not needed, they are bypassed.

Oct 30, 1993 This can reduce CPU and DASD requirements significantly.

If no INCODE=, no INTERVAL=, no SHIFT=, and no NORMX= are

used and there is only one input dataset, the first data

step is bypassed. If there is no SUMBY=, then no SORT.

Unfortunately, most of the ASUMxxxx and TRNDxxxx members

that invoke VMXGSUM do not meet the criteria and see no

benefit, but ANALDB2R gets some help, and it is the

righteous thing to do anyhow. See Change 11.309.

Change 11.235 Trending of VM/XA and VM/ESA MONWRITE dataset VMXAINTV

TRNDVMXA had logic errors, that causes several variables to be

Oct 30, 1993 incorrect.

Thanks to Sean Chaffee, Amadeus Data Processing GmbH & Co., GERMANY.

Change 11.234 Support for SMF type 42 subtype 14 (IBM ADSM Distributed

VMAC42 Storage Management Product) provides good statistics in

Oct 30, 1993 the new TYPE42AD dataset, with counts of byte traffic

and objects, for backup and archived files moved between

client and server, as well as CPU time plus waiting time

for Communications or Media, and session idle time.

The only identifiers, however, are the client name and

the client owner name. Unrelatedly, but because VMAC42

was "open", DFSMS Data Set, Storage Class, and Total

datasets were enhanced with read and write hit percents.

Thanks to Peggy Schulte, Cincinatti Gas and Electric Company, USA.
Change 11.233 SAS option SORTDEV=3380 was removed from MXG's CONFIG

CONFIG members - it is not needed, and caused errors at sites

Oct 30, 1993 that have no 3380s.

Thanks to Freddie Arie, Lone Star Gas, TEXAS.

Change 11.232 Variable AVGMTPTM was retained across device records that

VMAC74 had no Mount Pending. After its calculation, insert

Oct 30, 1993 ELSE AVGMTPTM=.;
Change 11.231 Additional enhancements and corrections to RMF reports.

ANALRMFR MXGCPU - Added OMVSMAX,OMVSAVG statistics.

Oct 30, 1993 MXGPAGE - Added OMVS00-OMVS11 statistics

MXGDEVC - DEVMTPTM to DEVACTTM, changed summary BY LCU

for calculation of AVGMTPTM.

MXGSMRY - BY SYSTEM DATE TIME changed to BY SYSTEM DATE.

Final page per system for overall total/avgs.

Corrected CPU busy percentages (only wrong if

you requested 24 hour report spanning 2 days).

MXGWKLD Added OMVS swap counts.

MXGENQU Circumvented "MISSING VALUE" note.

Thanks to Bill Stoneberg, National-Oilwell, USA

Change 11.230 MXG 11.06 or later. Landmark CICS causes "INVALID

TYPEMON8 ARGUMENT TO FUNCTION MDY" or "MXG HAS DETECTED INVALID

Dec 1, 1993 FORMAT FOR TIESDATE" because MO and DD were reversed in

the MDY() function. Change the MDY() function to read:

TIESDATE=MDY(MO,DD,YY);

See also change 11.278.

Thanks to Tom Clark, SEI Corporation, USA.

Thanks to Mark Robbins, W.H. Smith Limited, ENGLAND.

Change 11.229 Change 11.199 still did not calculate GMTOFF correctly.

VMAC30 Instead of the two IFs and kludgy FLOOR/CEIL logic, the

VMAC7072 correction is to use a single statement with +10 added:

Oct 30, 1993 GMTOFF70=100*FLOOR((STARTIME+DURATM-SYNCTIME+10)/100);

GMTOFF72=100*FLOOR((STARTIME+DURATM-SYNCTIME+10)/100);

GMTOFF30=100*FLOOR((SMF30IST-INTBTIME+10)/100);

Change 11.228 Open Edition/MVS (OMVS) dataset TYPE30OM was added to the

BUILDPDB PDB library. At present, it is simply sorted into the

BUILD005 PDB, but this logic may change with more experience with

Oct 30, 1993 the new data for "Unix under MVS".

Change 11.227 These new (calculated) variables were added to TYPE71,

TRND70 TRND71, and TRNDRMFI, as they have been found useful in

TRND71 MVS/ESA 4.3 analysis of paging and memory.

TRNDRMFI BLKPAGE ='BLOCKED*PAGES*PAGED IN'

VMAC71 NBLKPAGE='NON-BLOCKED*PAGES*PAGED IN'

VMAC7072 COMPAGE ='TOTAL*COMMON*PAGING RATE'

Oct 30, 1993 CS_ESPAG='TOTAL*CSTORE TO/FROM ESTORE*PAGING RATE'

HIPEPAGE='TOTAL*HIPERSPACE*PAGING RATE'

PGPERBLK='PAGES*PER*BLOCK'

SWAPAUX ='SWAP RATE*TO*AUXSTORE'

CSTORE ='CENTRAL*STORAGE*ONLINE'

ESTORE ='EXPANDED*STORAGE*ONLINE'

Variable PCTRDYWT was added in TRND70, and TRNDRMFI was

updated to include the ESA variables that had been added

in RMFINTRV but had been overlooked in TRNDRMFI.

Thanks to Diane Eppestine, Southwestern Bell, USA.

Change 11.226 JES2 NJE Purge Records for Job Transmission should have

BUILDPDB been OUTPUT to PDB.NJEPURGE, but were not; in rare cases,

BUILD005 they were used in place of the real JES2 Execution Purge

VMAC26J2 record. In BUILDPDB, the test to OUTPUT PDB.NJEPURGE was

Oct 29, 1993 expanded from (INREASON=' ' OR INREASON='JR' AND ... to

((INREASON=' ' OR INREASON='JR' OR INREASON='JT') AND ..

and VMAC26J2 logic was modified to test if INREASON was

still blank after testing INDEVICE, and if so, the same

decoding was applied to DEVNAME. The specific case that

was encountered had INDEVICE=INTRDR,DEVNAME=L329.JT1.

This change can alter the number of observations in the

PDB.JOBS and PDB.NJEPURGE datasets.

Thanks to Don Friesen, B.C. Systems, CANADA.
Change 11.225 Support for Amdahl APAF Version 2.1 adds "Effective" CPU

EXAPAFCH time, "Dispatched" CPU time, and the "In-Domain Mgmt" CPU

EXAPAFCP time variables for each MDF domain in dataset APAFDOMA.

IMACAPAF The "Outside-Domain Mgmt" CPU time, (IBM's "PHYSICAL")

VMACAPAF has been added to dataset APAFGLOB. New CPU Busy dataset

Oct 27, 1993 APAFCPBY provides, for each LP, the sampled percent CPU

busy in Problem state and in Supervisor state. Finally,

the new "Channel Busy" dataset APAFCHAN provides, for

each CHPID, separately for each Domain, the sampled CHAN

busy statistics. There's a lot of good stuff here! The

new subtype 4 and 5 code has only been syntax checked as

of this date, as those new data records were not provided

in the Amdahl sample set, but the APAFDOMA/APAFGLOB data

has been processed. Ultimately, there will probably need

to be a summary routine (like ASUMPRSM, will be named

ASUMAPAF if/when it exists), but only after test data and

some user experience is available! See Change 11.290.

Thanks to George Scott, Rockwell International Corporation, USA.

Change 11.224 CICS/ESA "Requested Reset Statistics", statistics records

CICINTRV with SMFSTRQT='RRT', were overlooked in member CICINTRV,

EXCICRRT but now new dataset PDB.CICRRTRV will be created from the

IMACCICS RRT records. Because this change adds a new dataset, if

Oct 25, 1993 you have tailored IMACCICS in your USERID.SOURCLIB, you

MUST retrofit your tailoring to the new IMACCICS, or you

will encounter "CICRRTRV NOT FOUND" syntax errors.

Thanks to Siegfried Trantes, Gothaer Versicherungsbank VVAG, GERMANY.

Change 11.223 INVALID DATA FOR OWNEXPDT message results when hex zeroes

VMAC6156 are in the Packed Decimal field, but there is no effect

Oct 25, 1993 on dataset TYPE6156, since SAS sets OWNEXPDT missing. By

inserting two question marks between OWNEXPDT and the

PD3. (or &PD.3.) informat, the invalid data message is

suppressed.

Thanks to Siegfried Trantes, Gothaer Versicherungsbank VVAG, GERMANY.
Change 11.222 Variable VIO occurs twice in the NORM1 argument in TRND71

TRND71 causing its value to be incorrect. Remove the first

Oct 25, 1993 occurrence. Variable CPUPAGTM (new in MVS/ESA 4.3) was

not kept in TRND71, so add it at the end of the SUM= .

Thanks to Diane Eppestine, Southwestern Bell.
Change 11.221 CICS/ESA excluded fields can be identified on the output

IMACEXCL of UTILCICS by the value of 65535 ('FF'x) for CMODOFST

UTILCICS (i.e., there is no offset to that field). A title was

Oct 25, 1993 added to that report documenting this fact. Comments in

IMACEXCL for the MACRO _CICXCUS definition now identify

which fields were added in CICS/ESA 3.2.1 and 3.3.

Thanks to Pierre Thibodeau, CGI Group, CANADA.
Change 11.220 Changes required only for execution under ASCII SAS:

VMACCMA a.SAS under Windows has a problem with old-style macros if

VMAC28 the contents start in column 1. Shifting the source code

VMAC102 one byte to the right circumvents the problem. This was

VMACLMS necessary in members VMACCMA, VMAC28, VMAC102.

ANALSMF b.SAS under Windows has a problem with ELSE ending one line

Oct 24, 1993 and its IF starting in column 1 of the next line. The IF

statements in VMACLMS were shifted right one column.

c.ANALSMF contained LENGTH 2, which was changed to 3.

d.Many ANAL members still had the old concatenation symbol

(pair of vertical bars, '4F'x on MVS) which is translated

to brackets under ASCII, which is not recognized by SAS

as concatenation. All were changed to exclamation point

which are acceptable to both ASCII and EBCDIC SAS.

Change 11.219 Support for FOCUS MSO 6.8 added variables FOCUACCT,

VMACFOCU FOCUCOMP,FOCUFACT,FOCUID1,FOCUID2,and FOCUPRTY to dataset

Oct 21, 1993 FOCUSMSO, and variable FOUCFRV no longer exists. Since

these fields were added compatibly to the SMF record,

previous MXG versions will process the new record (but

won't input the new variables!). Additionally, variables

SYSTEM and SMFTIME are now kept.

Thanks to Scott Ashby, Wachovia Operational Services Corp, USA.

Change 11.218 VM variable PWCOUNT is strangely stored by IBM. Values

TYPEVM are EBCDIC numbers 0 thru 9 (i.e., '40F0'x thru '40F9'x),

Oct 21, 1993 but a value of ten is stored as '0A', or 'F0C1'x, sixteen

is '0F' or 'F0C6'x, etc!). MXG now parses each byte and

figures out what IBM did so you get the right number.
Change 11.217 Division by zero can occur in processing TYPE78 data.

RMFINTRV The three lines with /NRATTMPS need to be placed in a DO

Oct 21, 1993 group : IF NRATTMPS GT 0 THEN DO; ... END; so that

division only occurs if NRATTMPS is greater than zero.

Thanks to Tom Miron, State of Wisconsin Dept of Administration, USA.
Change 11.216 GRAFTRND did not print workload data for any workloads

GRAFTRND after an unused workload. So if you don't use the IMS

Oct 21, 1993 workload, the TSO, BATCH, and all OTHx workloads do not

appear on the output graph. Change the 15 occurrences of

CPUSUM=CPUSUM+CPU; to read CPUSUM=SUM(CPUSUM,CPU,0);

Thanks to Bill Stoneberg, National Oilwell, USA.

Change 11.215 SMF type 57 variables added by Enhanced SYSOUT Support

VMAC57 (SMF57LN5,SGT,IND,RSV,JDT, TUL, and TU) are trashed for

Oct 21, 1993 sites without the ESS segment (they should be missing).

Change the test IF OFFESS NE 0 AND NRESS NE 0 THEN DO;

to read IF OFFESS GT 0 AND NRESS GT 0 THEN DO;

so the loop is executed only when the ESS section is in

the record. (NE is TRUE for a missing value, GT is not).

Thanks to John Mattson, National Medical Enterprises, USA.

Change 11.214 JES3 variable CLASS (Class on Main) was added to MACRO

IMACPDB _PDB26J3, so it will now be kept in the PDB.JOBS dataset.

Oct 21, 1993

Thanks to Hong-Suk Yang, Dept. of Health & Housing, AUSTRALIA

Change 11.213 Variable IOTM6421 in TYPE30 should have been formatted

VMAC30 as TIME12.2; now it is.

Oct 21, 1993

Thanks to Dan Squillace, SAS Institute Cary, USA

Change 11.212 ALIGN must be specified for the ASM of this VVDS-grazer

ASMVVDS program, as VMACVVDS expects the data record to contain

Oct 20, 1993 the extra bytes created by the ALIGN Specification. I

completely forgot that ALIGN/NOALIGN can affect the data

format for a DS field, and it appears that most sites use

ALIGN by default (since the program has been in use for

years), but one site whose default was NOALIGN had their

TYPEVVDS program ABEND, so the JCL now explicitly has the

ALIGN parameter.

Apologies to whoever found this problem, whose name/company I lost.

Change 11.211 CICS SAP Accounting variables STCDB1-STCDB5 are character

IMACICSA variables, not numbers. Find all occurrences of STCDB and

Oct 20, 1993 change their label from "CALLS" to "NAME" in five places

and change their INPUT from "&PIB.4." to "$CHAR4." in ten

places.

Thanks to Mr. Bodenbender, BUDERUS, GERMANY.

Change 11.210 The FACOM pseudo-RACF product's type 127 SMF record will

VMACF127 cause "FUNCTION CHAN IS UNKNOWN" and "VARIABLE CHAN HAS

Oct 13, 1993 ALREADY BEEN DEFINED" if the F127 processing is added to

BUILDPDB. Change "CHAN" to "CHANS" in the ARRAY, RETAIN,

and two occurrences of CHAN(CHANNUM+1) statements.

Thanks to Denis Johnco, Datacom Systems Ltd, NEW ZEALAND.

Change 11.209 The INPUT of variables MVTRERR,MVTWERR,MVPRERR,MVPWERR

VMACCOMP in VMACEDGS and variable ULOGCPUT in VMACCOMP need period