4. REPORTS › 4.3 Report Descriptions › FACILITY PROCEDURES
FACILITY PROCEDURES
The remainder of this chapter outlines some special
instructions and procedures for using certain of the CA MICS
StorageMate facilities. These instructions relate directly
to the use of some of the reports described in the first part
of this section.
INSTRUCTIONS FOR CREATING AND MAINTAINING THE PATTERN FILE
----------------------------------------------------------
One facility in CA MICS StorageMate allows you to create and
test various data set name patterns to be used in writing a
DFSMS ACS (Automatic Class Selection) Routine. If you plan
to use this facility, you must create a pattern file in your
Work Data Set that will be used as input by the facility.
Multiple pattern files can exist in your Work Data Set.
After its creation, the pattern file will probably continue
to be updated and refined as you run reports and add new
patterns to the file. The creation and maintenance of this
file is done external to CA MICS StorageMate, using an edit
facility such as ISPF Edit. You will create and maintain
these files using member name(s) of your own choosing. The
name you select must then be entered on the execution options
screen when you run the ACS Routine Definition report.
The syntax rules for the pattern file are as follows:
GENERAL SYNTAX RULES
Positions 1-72 of each input statement are scanned, with
positions 73-80 being ignored. We suggest you use these last
8 positions for sequence numbers.
COMMENT STATEMENTS
Comment statements are allowed and must have an asterisk
('*') as the first non-blank character, followed by at least
ONE BLANK.
THE PATTERN STATEMENT
The pattern statement contains up to three different items:
the pattern (required), the construct name (optional), and
the print switch (optional). These items can start in any
position, but must be separated by spaces, must appear in the
order listed above, and cannot contain any embedded spaces.
DATA SET NAME PATTERNS
The first value on each pattern statement consists of the
data set name pattern itself. Each pattern will be compared
against your data set names with the goal of finding a match
and assigning a construct name to the data set. Patterns
are tested in the order they are defined. Each pattern name
can be a maximum of 44 characters in length. Patterns
greater than eight characters must consist of qualifiers,
separated by periods (.), with not more than eight
characters per qualifier. Patterns containing none of the
special characters described in this paragraph are
considered fully-qualified data set names, and must match
exactly. Three special characters can be used. A %
(percent) represents a single character, and will match any
one position. An * (asterisk) within a qualifier represents
0-8 characters, or can also be used to represent a single
qualifier. A ** (double asterisk) represents zero or more
qualifiers, and cannot appear with any other characters in
the same qualifier. More than three adjacent asterisks is
considered an error. Refer to the IBM Manual "MVS/ESA
Storage Administration Reference" for more detailed
explanations of the syntax just described. Refer also to
the example later in this section.
THE CONSTRUCT
The pattern can, optionally, be followed by a 1-8 character
construct name. This name will be assigned to each data set
whose name matches the pattern defined previously on the
same statement. The construct name should consist of
alphabetic (A-Z) or numeric (0-9) characters, and should not
contain special characters or embedded blanks. The
construct name also cannot be the values PRINT or NOPRINT,
as those are reserved keywords. If no construct name is
specified, a value of "(NONE)" will be assigned.
THE PRINT SWITCH
The pattern or construct name can, optionally, be followed
by a keyword with two possible values: Specify PRINT to
receive a detailed report of each data set name that
matches this pattern or NOPRINT for no detailed report for
this pattern. NOPRINT is the default if this is left
blank.
UNMATCHED DATA SET STATEMENT
A third type of statement that can exist in the pattern file
is a keyword used to control the printing of data set names
that fail to match any pattern in the file. This keyword, if
coded, must exist as a separate statement in the file.
Specify UNMATCHED=PRINT to print data sets names that do not
match any pattern or UNMATCHED=NOPRINT to suppress printing.
No spaces should precede or follow the equal sign in the
keyword. The default is to not print unmatched data set
names. Specify this value only once in the pattern file. If
it occurs more than once, the last option coded will be used.
SAMPLE PATTERN FILE:
* SAMPLE PATTERN FILE FOR TESTING
UNMATCHED=PRINT
SYS%.** SYSPROG NOPRINT
**.*LIST WORKDSN NOPRINT
TEMP.** WORKDSN NOPRINT
PROD%.**.MASTER CACHED1 PRINT
IMS.**.BMP.PROD CACHED1 PRINT
SYS1.LINKLIB SPECIAL NOPRINT
%%%.CRIT.** BACKUP2 PRINT
SYS1.HASP* SYSPROG NOPRINT
PROCEDURES FOR USING THE ACS TEST FACILITY
------------------------------------------
IBM's ISMF online facility provides a mechanism for testing
the operation of your ACS routines. This is a useful
facility when you first create your ACS routines, and also
for ongoing maintenance. The test facility uses as input a
test case library, containing test cases (simulated
allocations of new data sets) that will be passed to your ACS
routines. The test case library is a Partitioned Data Set
(PDS), with each of the test cases existing as a separate
member of the library. The output of the test facility is a
report file containing the results of the testing for each
test case member. Figure 4-74 illustrates the use and data
flow of this facility.
+----------+ +----------+
| CA MICS | | Tape |
| Data | | Data |
+----------+ +----------+
| |
+------------+------------+
|
V
+---------------+
+--------+ | StorageMate |
| Report |<-----| ACS Test |
+--------+ | Case Creation |
+---------------+
|
|
V
+-----------+
+------------| Test Case |
| | Library |
| +-----------+
| |
V |
+--------------+ |
| IBM ISMF ACS | |
| Testing | |
| Facility | |
+--------------+ |
| |
| |
V |
+---------+ |
| Testing | |
| Results | |
+---------+ |
| V
| +--------------+
| | StorageMate |
+----------->| ACS Testing |
| Results |
+--------------+
| +---------+
| | |
+--------->| Reports |
| |
+---------+
Figure 4-87. ACS Test Facility Processing
While this facility is useful, building the test cases and
analyzing the output can be time consuming, especially if you
plan to use a large number of test cases. Two facilities
provided with CA MICS StorageMate can help in this area:
o The ACS Test Case Creation facility uses CA MICS data to
build a test case library and populate it with test cases.
o The ACS Testing Results facility reads the output file from
the testing facility and summarizes the test results.
You should use the following procedure when using these
facilities to test your ACS routines:
1) Use the ACS Test Case Creation facility to build test
cases from your CA MICS data. The output of the facility
is a jobstream in your Work Data Set.
2) Submit the jobstream created in the Work Data Set. It
will create and populate the test case library. One
member will exist in the library for each test case. The
member name will have the following structure:
@dddcnnn
where:
ddd = The first 3 characters of the data set
name being tested
c = A letter A-Z
nnn = A numeric value 0-999
A member named ZZZINDEX will also be created that will
contain information necessary to cross-reference the test
cases with the testing output file.
3) From the ISMF main menu, select the option for AUTOMATIC
CLASS SELECTION. From the next menu displayed, select
the TEST option. Then select TEST again from the next
menu.
4) On the test facility screen, you must specify:
o The name of the CDS library containing the ACS routines
to be tested.
o The name of the ACS test library. Specify the data set
name that was created in Step 2 above.
o The names of the numbers to be tested. Specify @*,
because this will test all members but the ZZZINDEX
member. You can also use this to subset the testing to
certain members. For example, you could specify @PAY*
to test only test cases for data sets with the prefix
'PAY'.
o The testing output report data set that will contain
the testing results.
o The combination of the four possible ACS routines to be
tested. Testing all four routines in the same pass is
possible, but will lengthen the time to complete the
test.
When you have invoked the test facility and the test
completes, the testing output report data set will be
displayed at the terminal. Press END to terminate
browse. You will then specify what to do with the
testing output report data set. Specify that the data
set be kept, but not printed.
5) Invoke the ACS Testing Results facility of CA MICS
StorageMate. As parameters to this facility, you must
supply the name of the test case library created in step
2 above, and also the testing output report data set
created in step 4 above. It is important the two data
sets specified represent the same data. The specified
testing output report data set must represent the testing
of the specified test case library. The ACS Test Results
facility can now be used to produce a number of reports
to evaluate the testing.
