IBM interview question ...Do reply pls.



IBM interview question ...Do reply pls.

Postby mr.ashutosh.kr » Wed Aug 11, 2010 11:21 pm

Question was:-----------
1.How would you test your load module ?
you are not left with the source code. you have only load module how would test that ?
2. Suppose for a jcl job i am getting maxcc = 0. now what i want you to display is maxcc = 1024 for the same jcl job ?
3. What is the difference between IGYWCLG and IGYCRCTL ?
4. How many no. of maximum columns is possible for creating a primary key for a table ?
5. What is the basic difference between FIXED BLOCK and VARIABLE BLOCK for a PS , PDS AND PDSE ?

What i answered:-----
1. LOAD MODULE---I DID NOT GIVE THE ANSWER.

2. IGYWCLG IS FOR COMPILING,LINKING-EDITING AND GO.
IGYCRCTL IS FOR JUST COMPILING.

3. MAXCC--------DID NOT ANSWERED

4. COMBINATION OF ENTIRE COLUMN PRESENT IN A TABLE CAN BE USED TO DEFINE A PRIMARY KEY I.E. FOR A TABLE MAX NO. COLUMN POSSIBLE IS 750.

5.its a kind of record format used in jcl to provide the information how the data is going to be allocate.

in fixed block the fix space would be allocated for the data and in variable depends upon the size of data it would allocate the space.
mr.ashutosh.kr
 
Posts: 2
Joined: Wed Aug 11, 2010 10:37 pm
Has thanked: 0 time
Been thanked: 0 time

Re: IBM interview question ...Do reply pls.

Postby Robert Sample » Wed Aug 11, 2010 11:56 pm

1. This is REAL black-box testing! Look at the specifications for the program and provide inputs that match the specifications to ensure the program is working correctly for good input, then provide outputs that do not match the specifications to verify that the program handles incorrect input correctly. Then you can try different things to see what happens, for example, if the program runs with no DD statements, to check on how it handles wild card behavior.

2. I would ask why someone would want a zero job return code changed to 1024 -- this makes no sense. However, if required I would add a step running a program I wrote in COBOL to set the RETURN-CODE to 1024. This step would be bypassed if any previous step had a non-zero return code.

3. IGYWCLG is JCL provided by IBM to compile, link, and execute (run) a COBOL program. IGYCRCTL is the name of the COBOL compiler for Enterprise COBOL and is executed in the first step of IGYWCLG.

4. Assuming this is DB2, I don't deal with DB2 so couldn't answer.

5. Fixed block means each block of data (except possibly the last one) will contain the same number of bytes. The LRECL must evenly divide the block size (you cannot put 6.5 records in a block). Variable block means the number of bytes in each block can vary. LRECL does not evenly divide the block size. Each block and each record has a 4-byte prefix that contains the block or record length in it. There's not really any difference in PS and PDS files as far as block size goes; I don't know about PDSE as I don't use them enough.
Robert Sample
Global moderator
 
Posts: 3719
Joined: Sat Dec 19, 2009 8:32 pm
Location: Dubuque, Iowa, USA
Has thanked: 1 time
Been thanked: 279 times

Re: IBM interview question ...Do reply pls.

Postby steve-myers » Thu Aug 12, 2010 3:56 am

Actually, Mr. Sample's discussion of fixed block is incorrect. Mr. Samples is describing fixed block standard (FBS). The only requirement for FB is each block is a multiple of the LRECL, but there is no requirement for any block to be the maximum length block.

As an experiment a few years ago I wrote a PDS copy utility for FB and VB data sets where I wrote for the last block on a track the maximum block that would actually fit and was consistent with the LRECL rule. I had high hopes this would save a lot of space, but it didn't, even though each track was optimzed. However, SPF and other programs read this weird stuff without complaint.
steve-myers
Global moderator
 
Posts: 2105
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 243 times

Re: IBM interview question ...Do reply pls.

Postby dick scherrer » Thu Aug 12, 2010 4:10 am

Hello,

The only requirement for FB is each block is a multiple of the LRECL, but there is no requirement for any block to be the maximum length block.

Might not be a "requirement", but you'll not find typical qsam files that were written with "short blocks" in the middle - all but the last block will be blksize . . . Things can happen later to cause embedded "short blocks" but this is not an interview question for a student or "fresher"/beginner.
Hope this helps,
d.sch.
User avatar
dick scherrer
Global moderator
 
Posts: 6268
Joined: Sat Jun 09, 2007 8:58 am
Has thanked: 3 times
Been thanked: 93 times

Re: IBM interview question ...Do reply pls.

Postby steve-myers » Thu Aug 12, 2010 8:57 am

I agree. In any event, in the real world the only way you have short blocks embedded in an FB data set is if you do MOD a few times. This is why MOD is not allowed for FBS. My PO copy program that I discussed was cheating all the way, but it was a useful experiment, even if it didn't produce the results I had hoped for.
steve-myers
Global moderator
 
Posts: 2105
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 243 times

Re: IBM interview question ...Do reply pls.

Postby dick scherrer » Thu Aug 12, 2010 9:00 am

Hi Steve,

but it was a useful experiment, even if it didn't produce the results I had hoped for.
These are often the ones where the most is learned :)

Be well,

d
User avatar
dick scherrer
Global moderator
 
Posts: 6268
Joined: Sat Jun 09, 2007 8:58 am
Has thanked: 3 times
Been thanked: 93 times

Re: IBM interview question ...Do reply pls.

Postby mr.ashutosh.kr » Thu Aug 12, 2010 9:44 pm

Thank you for your answers.
mr.ashutosh.kr
 
Posts: 2
Joined: Wed Aug 11, 2010 10:37 pm
Has thanked: 0 time
Been thanked: 0 time


Return to Interview Questions

 


  • Related topics
    Replies
    Views
    Last post