Thread iteration limit value significance in concurrency thread group.

[K]

Hi Team,
I have one question,
Scenario: I have one scenario where I have almost 30k records in csv and I have to execute a query such that each unique user accesses the csv individually, i.e control their own cursor. 
E.g let's say I started with one user it will access data in csv and make requests, But when after let's say 2 mins, New user comes he should access the csv from start so both threads are accessing csv independently.

I know I can control the csv data set config to current thread, But unsure about the value of thread iteration limit.
Should I set thread iteration limit to 30k so each thread will execute 30k Iterations hence 30k records from csv?

Thanks

[Vincent Daburon]

Hi,

A solution is to use the Http Simple Table Server (STS) and load your 30k record csv file in the STS

Then vusers will ask a line to the STS and consume (remove) the line.

When a vuser iterate may 1000 times, the vuser stop (but other vusers continue).

When all values are remove (the file in the STS is now empty) all vusers stop

https://jmeter-plugins.org/wiki/HttpSimpleTableServer/

Regards

Vincent DAB.

[Dmitri T]

With default sharing mode of CSV Data Set Config of "All threads" each virtual user reads next file from the file on each iteration so as long as number of users * number of iterations < 30k the requests will be "unique" with default CSV Data Set Config setup. You can set "Stop thread on EOF?" to true to ensure that no duplicate requests will be sent.

On Friday, May 31, 2024 at 3:11:30 PM UTC+2 K wrote: