|
View previous topic :: View next topic
|
| Author |
Message |
Huzefa
New User
Joined: 05 Dec 2006
Posts: 83
Location: Bangalore
|
|
|
|
Hi
I am getting a SOC4 abend when I try to move the 516th record to a Working Storage variable.
I ahve the program compiled with 'NODYNAM,DATA(31) Option. I have the 'REGION=0M' specified on my JobCard. I have tried all possible options in order to avoid this error. There is no problem with the file I have checked and validated it.
One point that I notice is that it reads and displays every record properly in the spool. But when it reads and Displays the 516th record it displays only the first 102 bytes of the 213 bytes. Not sure why.
Below is the error log from the job
| Code: |
IEA995I SYMPTOM DUMP OUTPUT 127
USER COMPLETION CODE=4039 REASON CODE=00000000
TIME=15.43.23 SEQ=16944 CPU=0000 ASID=0295
PSW AT TIME OF ERROR 078D1000 88281CB2 ILC 2 INTC 0D
NO ACTIVE MODULE FOUND
NAME=UNKNOWN
DATA AT PSW 08281CAC - 00181610 0A0D58D0 D00498EC
AR/GR 0: A55BB58A/84000000 1: 00000000/84000FC7
2: 00000000/00028078 3: 00000000/00000004
4: 00000000/08276BB0 5: 00000000/00027C0C
6: 00000000/00000000 7: 00000000/2930B410
8: 00000000/00028078 9: 00000000/0002977E
A: 00000000/00027C0C B: 00000000/88281BD8
C: 00000000/00021B88 D: 00000000/0002A878
E: 00000000/88275D74 F: 00000000/00000000
END OF SYMPTOM DUMP
+ABONL-23: UNABLE TO DETERMINE REPORT DATASET FILE TYPE |
Please let me know of any options that I can use to revert from the problem. |
|
| Back to top |
|
 |
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
| why not print ( using idcams for example ) 520 records to se what is going on in the dataset ??? |
|
| Back to top |
|
|
Huzefa
New User
Joined: 05 Dec 2006
Posts: 83
Location: Bangalore
|
|
|
|
I tried printing all the records and thats the reason I am sure there is nothing wrong in the records.
Infact I tried swapping the records i.e. Moved the first record that processed successfully to 516th record, but it still abended at 516th record.
Just FYI the file is a VB File |
|
| Back to top |
|
|
Bill O'Boyle
CICS Moderator
Joined: 14 Jan 2008
Posts: 2501
Location: Atlanta, Georgia, USA
|
|
|
|
You should specify DYNAM as opposed to NODYNAM. This is preferable for Batch but NEVER for CICS.
Are you using the following FD example -
| Code: |
FD INPUT-FILE
LABEL RECORDS ARE STANDARD
RECORDING MODE IS V
RECORD IS VARYING IN SIZE
FROM xxxxx TO yyyyy CHARACTERS
DEPENDING ON WS-INPUT-LGTH
BLOCK CONTAINS 0 RECORDS.
01 INPUT-REC PIC X(yyyyy).
WORKING-STORAGE SECTION.
01 WS-DATA.
03 WS-INPUT-LGTH PIC 9(08) COMP.
03 WS-INPUT-REC PIC X(yyyyy).
PROCEDURE DIVISION.
READ INPUT-FILE INTO WS-INPUT-REC.
|
After the READ, WS-INPUT-LGTH will contain the length of the variable-length record which was just read and INPUT-REC AND WS-INPUT-REC must be addressed using reference modification with WS-INPUT-LGTH as the length portion.
WS-INPUT-REC (1:WS-INPUT-LGTH).
A S0C4 is certainly possible if you attempt to address INPUT-REC without using reference modification and (for example) WS-INPUT-LGTH was 100 and you tried to address a position in INPUT-REC beyond position 100.
Note in the example, xxxxx represents the minimum record-length and yyyyy represents the maximum record-length.
Bill |
|
| Back to top |
|
|
Huzefa
New User
Joined: 05 Dec 2006
Posts: 83
Location: Bangalore
|
|
|
|
The FD Section is the exact same as what is mentioned
I am getting a SOC-4 when I execute the move as below for 216th record
| Code: |
| MOVE INPUT-REC to WS-INPUT-REC |
I feel the problem is somewhere related to the memory overflow or something but I am not very sure or know the resolution. |
|
| Back to top |
|
|
Bill O'Boyle
CICS Moderator
Joined: 14 Jan 2008
Posts: 2501
Location: Atlanta, Georgia, USA
|
|
|
|
I would suggest using the example I had posted and use reference modification, which was introduced with VS/COBOL II. If you're using OS/VS COBOL, then this won't work.
When you READ a variable-length file and the FD "01" is defined as 213 bytes, but the actual length for a given READ is less than 213 bytes (let's say it's 200), then in your MOVE statement, the underlying Assembler instruction will attempt to move all 213 bytes from INPUT-REC to WS-INPUT-REC. When the instruction attempts to move byte 201 from INPUT-REC to WS-INPUT-REC, you'll get a S0C4 addressability (protection) exception because the storage allocation after the READ for "INPUT-REC" was only 200 bytes and the address for byte 201 and greater is bogus.
I can't explain this any clearer so, it's your call....
Bill |
|
| Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8696
Location: Dubuque, Iowa, USA
|
|
|
|
Huzefa, it sounds like you've decided what the problem is and want a solution to that problem, whether or not it actually has any bearing upon your actual problem. Having the right attitude is 90% of debugging, and the right attitude is to not decide in advance what the problem is.
What is the file status code for the read of the record you think is having the problem?
What is the VSAM extended file status codes for the read of that record?
If the file is variable length, coding MOVE INPUT-REC TO WS-INPUT-REC is probably the single worst way to move the buffer data to WORKING-STORAGE. You need to either use reference modification to do the move, or you need one 01 level in the FD for each possible length record and move the correct 01 level to WORKING-STORAGE. Even better would be to use READ INTO.
And a protection exception (S0C4) may or may not have anything to do with a file -- it can be caused by different things, some of which have nothing to do with files. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
A couple of thoughts. . .
Is there an array involved?
Does the file have a "depending on some-lth-field"?
What happens if you define a ws variable that is 40,000 bytes and "read into" this rather than using the MOVE?
| Quote: |
| I feel the problem is somewhere related to the memory overflow or something but I am not very sure or know the resolution. |
It would be better if you changed feelings. . . Your feeling should be to look for the problem with either the file definition, the file content, or (most likely) the code is wrong. . . |
|
| Back to top |
|
|
Huzefa
New User
Joined: 05 Dec 2006
Posts: 83
Location: Bangalore
|
|
|
|
I tried Reference Modification and it doesn't work.
Below is the code snippet to answer all the questions.
| Code: |
FD INPT-FILE
RECORDING MODE IS V
RECORD IS VARYING IN SIZE
FROM 1 TO 250 CHARACTERS
DEPENDING ON INPT-LEN
DATA RECORDS ARE INBOUND-INPT-AREA.
01 INBOUND-INPT-AREA.
05 I-INPT-REC-TYPE PIC 9(02).
05 I-INPT-REC-DATE PIC 9(08).
05 FILLER PIC X(240).
01 RECORD-SPACE.
05 REC-LEN PIC S9(04) COMP.
05 REC-RETURNED.
10 REC-TYPE PIC X(02).
10 REC-CPCS-DT PIC 9(08).
10 REC-RETRN-BALANCE PIC X(240).
READ INPT-FILE
AT END
SET EOF TO TRUE
END-READ
EVALUATE INPT-STAT
WHEN 00
WHEN 04
DISPLAY 'WHEN 0'
MOVE INPT-LEN TO REC-LEN
DISPLAY 'INBOUND-RTRN-AREA ' INBOUND-RTRN-AREA
MOVE INBOUND-INPT-AREA(1:REC-LEN) TO REC-RETURNED
DISPLAY 'REC-RETURNED ' REC-RETURNED
WHEN 10
SET END-OF-PROCESS TO TRUE
WHEN OTHER
SET END-OF-PROCESS TO TRUE
PERFORM ABEND-PARA
END-EVALUATE |
Please let me know if I have missed anything. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
| Quote: |
| Please let me know if I have missed anything. |
Yes - for starters the FD does not match the 2 level-01s.
You didn't use "read into" as suggested.
Where is "inpt-len"?
Where is inpt-stst?
 |
|
| Back to top |
|
|
ridgewalker58
New User
Joined: 26 Sep 2008
Posts: 51
Location: New York
|
|
|
|
I have used the following structure and it works.
FD INPUT-FILE
LABEL RECORDS ARE STANDARD
RECORDING MODE IS V
RECORD IS VARYING IN SIZE
FROM 20 TO 1500 CHARACTERS
DEPENDING ON WS-INPUT-RECSIZE
BLOCK CONTAINS 0 RECORDS.
01 INPUT-REC-ALL.
05 INPUT-REC-BYTES OCCURS 20 TO 1500
DEPENDING ON WS-INPUT-RECSIZE PIC X(001).
WORKING-STORAGE SECTION.
01 WS-INPUT-RECSIZE PIC 9(08) COMP VALUE ZERO.
01 WS-INPUT-RECORD PIC X(1500) VALUE SPACES.
PROCEDURE DIVISION.
READ INPUT-FILE
AT END
MOVE 'Y' TO INPUT-FILE-EOF.
IF INPUT-FILE-EOF = 'Y'
GO TO Q0000-EXIT.
MOVE INPUT-REC-ALL TO WS-INPUT-RECORD. |
|
| Back to top |
|
|
PalDesh22
New User
Joined: 11 Nov 2010
Posts: 2
Location: India
|
|
|
|
| Why dont you change the input file to have only 515 records and check if that runs. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello and welcome to the forum,
When replying to a topic, it is best to read and understand the posts leading up to the current "last" reply.
Did you read/understand this:
| Quote: |
| Infact I tried swapping the records i.e. Moved the first record that processed successfully to 516th record, but it still abended at 516th record. |
The problem does not appear to be related to the content of record 516. . .
If the program runs to end with no abend using only 515, what might be learned?
I suspect the problem progrm is "walkikng on storage" and the problem is not the actual abend, but the processing that occurs before the abend. |
|
| Back to top |
|
|
Huzefa
New User
Joined: 05 Dec 2006
Posts: 83
Location: Bangalore
|
|
|
|
Finally!!!
I modified the READ to READ INTo and the program worked like a charm.
Thanks all for your patience and Reply!!!! |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Good to hear it is working - thank you for letting us know 
Now that it is working, i'd encourage spending some time to learn exactly why there was a problem.
Good luck,
d |
|
| Back to top |
|
|
|
|
