The SMF data set dump program creates a summary activity report when the DUMP option was
specified for any of the input data sets and at least one record was read or written.
Figure 1 shows an example of the summary activity report when
you specify the REPORTOPTS(NOSUBTYPE) parameter or allow it to default.
Figure 1. Example of summary activity report (without subtypes)
SUMMARY ACTIVITY REPORT
START DATE-TIME 06/17/2019-13:57:42 END DATE-TIME 06/17/2019-14:15:36
RECORD RECORDS PERCENT AVG. RECORD MIN. RECORD MAX. RECORD RECORDS
TYPE READ OF TOTAL LENGTH LENGTH LENGTH WRITTEN
2 296 47.36 % 605.34 18 612 297
3 1 .16 % 18.00 18 18 2
14 52 8.32 % 432.00 432 432 52
15 48 7.68 % 432.00 432 432 48
17 10 1.60 % 100.00 100 100 10
30 44 7.04 % 1,148.20 480 1,570 44
42 173 27.68 % 399.43 270 484 173
84 1 .16 % 1,344.00 1,344 1,344 1
TOTAL 625 100 % 550.98 18 1,570 627
NUMBER OF RECORDS IN ERROR 0
Figure 2 shows an example of the summary activity report when
specify the REPORTOPTS(SUBTYPE) parameter. If the record header indicates that subtypes are not
valid for a given record type, a dash (-) appears in the SUBTYPE column for that
record type.Figure 2. Example of summary activity report (with subtypes)
START DATE-TIME indicates the date and time of the earliest record read,
excluding record types 2, 3, and those greater than 127. Under most circumstances this date and time
will not equal the date and time of the first record written.
END DATE-TIME indicates the date and time of the latest record read,
excluding record types 2, 3, and those greater than 127. Under most circumstances this date and time
will not equal the date and time of the last record written.
RECORD TYPE indicates the identifying number of each record type read by
the SMF dump program.
RECORD SUBTYPE indicates the identifying number of the
subtype of each record type read by the SMF dump program when you specify the REPORTOPTS(SUBTYPE)
parameter.
RECORDS READ indicates the number of input records read for each record
type.
PERCENT OF TOTAL indicates the number of records read for each type
divided by the total number of records read. A value of
*** indicates that the percent could not be calculated because the
total number of records read exceeds
99,999,999,999.
AVG RECORD LENGTH, MIN RECORD LENGTH, and
MAX RECORD LENGTH indicate, respectively, the average, minimum, and maximum
lengths of the records read for each record type.
RECORDS WRITTEN indicates the number of output records written to the
output data sets as specified by the SYSIN parameters.
TOTAL indicates the total activity for each column.
NUMBER OF RECORDS IN ERROR indicates the number of records that were not
dumped because of errors. There are two types of errors:
Incorrectly-spanned records cannot be read. (A spanned record is larger than the
control-interval size of the SMF data set.) These records may be encountered if a physical I/O error
occurred on a recording data set while a spanned record was being written. The SMF data set dump
program is unable to read incorrectly-spanned records.
Incorrectly-formatted time or date values in record-header fields prevent processing of the
record. The record is read but not written or processed. If your installation writes user records
(record types ranging from 128 through 255), see Standard and Extended SMF record headers for a description of the
proper header format.
If there are any records in error, the INDD-name or LSNAME-name, type of error (INVALID TIME
OR DATE or INCONSISTENT SPANNED RECORD), record sequence number, and the first 32 bytes of the
record (in error) are printed. This processing is performed only for the first 500 records in error
of each SMF recording data set. When more than 500 records in error are found, the SMF data set dump
program writes message IFA024I to the SYSPRINT data set. Subsequent records in error are not
printed, but are counted in the total number of records in error. The record sequence number is the
sequential number of the record in the INDD.
Figure 3 shows an example of a record in error. Figure 3. Example of a Record in Error
INDD: INDD1
ERROR TYPE: INCONSISTENT SPANNED RECORD
RECORD SEQUENCE NUMBER: 5
005B0000 1E140056 51510092 060FF3F0 F9F0D4E2 E3D1C3D3 F0F00056 4D840092 * j 3090MSTJCL00 *
ERROR TYPE: INVALID TIME OR DATE
RECORD SEQUENCE NUMBER: 1,555,988
021D0000 1E3C0000 00000000 000003F0 F9F04040 4040E4D7 00000028 000A0001 * j 3090 UP *
INDD: INDD2
ERROR TYPE: INCONSISTENT SPANNED RECORD
RECORD SEQUENCE NUMBER: 8
010A0000 DE1E0056 51510092 060FF3F0 F9F0E2E3 C3400001 00000070 00160001 * j 3090STC *
When using IFASMFDL to process digitally
signed SMF records, see Using IFASMFDL to carry signatures to data sets for information about how signature data is
included in the summary report.
Note: When running z/OS V2R1 and V2R2, records that include SMF extended header information appear
as Type 126 records in the report. Because the dump utilities in releases V2R1 and V2R2 cannot write
extended header records, the RECORDS READ count is non-zero and the RECORDS WRITTEN count is zero.
When records with extended headers or record types larger than 255 characters are encountered, the
following line is added to the summary activity report.
when
you specify the REPORTOPTS(NOSUBTYPE) parameter or allow it to default
.