Monday, 2 April 2012

Loadrunner Interview Questions - Part IV


 Loadrunner Interview Questions - Part IV

1.How many types of Correlation are there?
 Correlation can be done by 3 ways: 

  1. Manual Correlation ( this method is most recommended)  
  2. Automatic Correlation (This requires rules need to be created in existing rule engine, this can apply only to popular and frequent dynamic values like viewstates in .Net applications) 
  3. Correlation Studio

2.How you will do the manual Correlation?
  1. Record two identical scripts with the same business process, make sure to use same input data like user names and search strings
  2. Compare the scripts to identify the dynamic value, You can compare them using notepad or keeping side by side or using windiff or any text comparison tools. the mainly you have to look at ITEMDATA section and URL ending part. Once you identified which value is changing every time
  3. Search for the dynamic values in the Generation log and Copy Left boundary (LB) and Right boundary (RB) into notepad
  4. Replay the script with extended log and look for those boundaries in replay log and Double click in the replay log  where the  boundaries are found and it will take the cursor to the point where dynamic value is generating in the recorded script. This step is to identify from which request this value is generating and this will help us to identify the location to keep web_reg_save_param function.
  5. Insert web_reg_save_param("Param name", "LB=" , "RB=", LAST) ; where the cursor is pointing       
  6.  Replace the dynamic values with Param name
3.How you will do the parameterization for Date/Time?
Select the parameter type as Date/Time  and select the date/time format in the list.
If you want u can get the date in the script by using "lr_save_datetime("Tomorrow is %B %d %Y",     DATE_NOW + ONE_DAY, "next"); "

4.How many areas you can do the correlation?
Areas of correlation are:
1) ItemData
2) TimeStamp
3) Links
4) Check Boxes
5) List Butoons
6) Radio Buttons

5.What is the function used for to find out the buffer size of file ?
First of all we have to know the application type for this question.
If it is web application we dont need to go for buffer size bcoz most of the functions are inbuilt with the buffer element.
If we go for windows sockets application the function is mentioned below:
char *lrs_get_static_buffer( char *s_desc, char *buffer, int offset, int length, char *encoding );

6.What is the difference between Hits/secs & Throughput?
Hits Per Second: This graph shows the number of HTTP requests made by Vusers to the Webserver during each second of the load test scenario run.
Throughput:This graph shows the amount of throughput on the server during each second of the load test scenario run. Throughput is measured in bytes or megabytes and represents the amount of data that the Vusers received from the server at any given second.

7.How graph will be there between Hits/secs & Throughput?
 
Hits per second is directly propositional to through put
 If the Hits/secs Increses Throughput should also increase
 if the Hits/sec decreases through put  should also decrease

8.If suppose we are executing a scenario with 10 vusers and a value is parameterized (Say ex: Sales Id with 10values), how can i know that what is the value of 5th vuser sales id?
The question is having little information to answer, Question should also have which parameter properties this scenario is using. Let us go with some examples. If the setting is sales id picked Sequential and update for each iteration. in the first iteration all the users will use first value, so the answer is 5th iteration. If it is Unique the 5th sales id will the 5th user will take. To find 5th vuser value we have to know first what we have give the  parameter properties.
If we take sequential-->each iteration then in parameter values 5th value is 5th vuser value and if we take unique also same value and if we take random we have to check in the log file of LRR folder.

9.What is the purpose of the Unique value?
Assigns a unique sequential value to the parameter for each Vuser. Ensure that there is enough data in the table for all Vusers and their iterations. Ifyou have 25 Vusers and you want to perform 4 iterations, your table must contain at least 100 unique values.
If you run out of unique values, VuGen behaves according to the option you select in the When run out of values field.

In real life applications, in lot of situations application will not accept duplicate values like order number, registration functionality. 

10.In controller, when I am executing the scenario with 10vusers, among 10vusers 1 is failed then how you will find out who is the exact user? 
First of all enable the ip spoofing and insert "lr_get_vuser_ip()" then execute the scenario and take the log files saved in result directory(.LRR extension) and find the failed user transaction then we can find vuser ip which is failed.

11.Is there any function in lr to avoid multiple login's if a user already logged in? 
There is no function to avoid multiple logins. In case if we have to avoid multiple login based on our application then we have to write our own code in "C" language.

12.In controller I am executing a scenario with 2Vusers and a value is paramerterized ( Say ex: Sales with 10values). So, whether it is possible to assign for each users 5 values? If yes how ? 
Go to parameter properties  
Selext Next Row:Unique 
Update each value on: Each occurence  
then automatically radio button " Allocate ________ values for       each vuser" will be enabeld

14 comments:

  1. Good Questions for interview purpose.

    Thanks For posting this type of questions. Plz Post some other use full information related to Load runner.

    ReplyDelete
  2. Thanks rams, I am mainly covering Loadrunnner Basics, once it is done I will focus on articles on. Performance engineering.

    ReplyDelete
  3. Good Question.Thanks for posting LR questions

    ReplyDelete