JMeter Beginner Tutorials | Tips n Tricks 8 💡 Concurrency Thread Group-How to setup Realistic Test
Вставка
- Опубліковано 25 лют 2018
- Free tutorials - automationstepbystep.com/
Hi, I am Raghav & Today we will learn:
1. How to use Concurrency Thread Group
2. How to control threads
References:
Stepping Thread Group is deprecated
jmeter-plugins.org/wiki/Stepp...
jmeter-plugins.org/wiki/Concu...
________________________________________________________________
Do like, share and subscribe
Keep Learning,
Raghav
_________________________________________________________________
Learn JMeter step-by-step for Beginners
JMeter Beginners - • JMeter Beginner
JMeter Intermediate - • Jmeter Intermediate
JMeter Advanced - • JMeter Advanced
JMeter Tips n Tricks - • JMeter Tips n Tricks
All Playlists - / @raghavpal
_______ ONLINE COURSES TO LEARN _______
automationstepbystep.com/909-...
----------- UI TESTING -----------
Selenium Beginners - bit.ly/2MGRS8K
Selenium Java Framework from Scratch - bit.ly/2N9xvR6
Selenium Python - bit.ly/2oyMp5x
Selenium Tips - bit.ly/2owxc50
Selenium Builder - bit.ly/2MKNtlq
Katalon Studio - bit.ly/2wARFdi
Robot Framework - bit.ly/2Px6Ue9
----------- API TESTING -----------
Web Services (API) - bit.ly/2MGafL7
SoapUI - bit.ly/2MGahmd
Postman - bit.ly/2wz8LrW
General - bit.ly/2PYdwmV
----------- MOBILE TESTING -----------
Mobile Playlist - bit.ly/2PxpeUv
----------- CI | CD | DEVOPS -----------
Jenkins Beginner - bit.ly/2MIn8EC
Jenkins Tips & Trick - bit.ly/2LRt6xC
Docker - bit.ly/2MInnzx
----------- VERSION CONTROL SYSTEM -----------
Git & GitHub - bit.ly/2Q1pagY
----------- PERFORMANCE TESTING -----------
JMeter Beginner - bit.ly/2oBbtIU
JMeter Intermediate - bit.ly/2oziNVB
JMeter Advanced - bit.ly/2Q22Y6a
JMeter Tips & Tricks - bit.ly/2NOfWD2
Performance Testing - bit.ly/2wEXbLS
----------- JAVA -----------
Java Beginners - bit.ly/2PVUcXs
Java Tips & Tricks - bit.ly/2CdcDnJ
----------- MAVEN -----------
Maven - bit.ly/2NJdDRS
----------- OTHERS -----------
Redis- bit.ly/2N9jyCG
Misc - bit.ly/2Q2q5xQ
Tools & Tips - bit.ly/2oBfwoR
QnA Friday- bit.ly/2NgwGpw
Sunday Special - bit.ly/2wB23BO
Ask Raghav - bit.ly/2CoJGWf
Interviews - bit.ly/2NIPPxk
All Playlists - bit.ly/2LSiezA
Keep Learning,
Raghav
Website - automationstepbystep.com/
LifeCharger - lifecharger.org/
Udemy Courses - www.udemy.com/user/raghav-pal-3/
Facebook - / automationstepbystep
Twitter - / automationsbs
UA-cam - / automationstepbystep
Online courses - automationstepbystep.com/
Instablaster.
Hi Raghav,
Thanks for the creating awesome tutorial series.I am a strict follower of your channel.Its very rare to find tutorial about the new features of Jmeter 4.I came across that we can also create HTML reports by executing using ant.It would be really helpful if you can make a video on that one .I tried from my end but coould not generate it for Jmeter 4.(java-1.8.0,ant 1.8).Please guide for HTML report generation using ant for Jmeter 4.
Thanks in Advance
Sure Swayam, will try to do it soon.
Hello, thank you for the videos.
I don't understand one thing, I made my configuration as: 20/20/5/5. (seconds selected)
And I got:
- 173 samples according to my "View Result in Table" (Wasn't it supposed to be 20?)
- 83% error rate according to "Aggregate Report".
The erros , according to my "View Results Tree" are:
- Response code: 503
It happens even testing google.com.
Is this plugin unstable nowadays? Thank you.
Hi Denis, I will need to check this as I get some time. Meanwhile you can do some more tests and verify the results
hi raghav I have a doubt that in the stepping thread group in which all threads did not shut down at once in this one, all threads shut down at once, so it does not seems to be realistic approach as all threads should shut down gradually as it happens with stepping thread group, is there any type of thread group in which there can be realistic approach as it was there in stepping thread group ??
Hi Keshav
The Concurrency Thread Group doesn’t provide the ability to define initial delay, step transition and ramp-down - which the Stepping Thread Group does.
read it here www.blazemeter.com/blog/advanced-load-testing-scenarios-jmeter-part-4-stepping-thread-group-and-concurrency-thread?BM_blog&
I have a question. I have concurrency thread group with 6 HTTP requests under parallel controller. I am running them non-gui. The result I am currently getting is that each of the 6 requests has graphs and which is correct. What I want now is that split the 6 HTTP request into two groups. Group A and Group B but still under the same parallel controller and run on non-gui. I want to have the same 6 graphs with the addition of the graph of them as a group. I want to see 8 graphs/results in total. How can I do that?
Hi Vincent, not sure, will need to check online
Hi Raghav. i am evaluating apache bench tool and JMETER. Problem is that for same concurrency params, if i run via AB, i get a very high RPS like 50K+ RPS, and when same thing if i run via JMETER concurrency thread group, i get only 1000 RPS. What am i missing ?
Hi Nabil, will need to check the setup. Can try some online help on this
Hi Raghav, I need to perform stress/load testing of the application using jmeter, where i want to see the behaviour of application for a 10,000 requests/ every 30 mins for a period of 5 hours. Also, i need to run this in non-gui mode.. How can i achieve it? Looking forward for your guidance on the same
HI Praveen, you will find a video on running in non gui, cmd in the JMeter playlist and you can set the configuration too. Check the JMeter all playlists here - automationstepbystep.com/
Hi,
When I try to use concurrency thread group I am facing stack overflow exeception:null , I am hitting the request and then moving along the pages by doing pagination and while controller, Can someone help me understand why am I getting the stack overflow exception, I checked my while controller as well it is not called infinitely.
Abhishek
The stack overflow error in JMeter typically occurs when there's an excessive amount of data being processed or a recursive function call that doesn't terminate. In the context of a Concurrency Thread Group, it's possible that the issue arises due to the combination of multiple threads executing the same code path and accumulating data within a limited memory space.
To effectively troubleshoot this issue, it's crucial to examine the specific code segments that might be causing the stack overflow. Here's a step-by-step approach:
1. **Identify the Code Path:** Determine the code path where the stack overflow error occurs. This could be a specific test script, a custom function, or a complex logic within your test plan.
2. **Check for Recursion:** Analyze the code for any recursive function calls. Recursive functions can lead to stack overflow if they don't have a clear termination condition.
3. **Review Data Processing:** Examine any data processing operations that might be accumulating data within the code. Ensure that data is being handled efficiently and not growing unboundedly.
4. **Validate While Controller Usage:** Verify that the While Controller is properly configured and doesn't lead to an infinite loop. Ensure the condition for exiting the loop is met and doesn't depend on external factors that might not change.
5. **Monitor Memory Usage:** Use JMeter's monitoring tools to observe the memory consumption during test execution. If the memory usage spikes excessively, it could indicate a memory leak or excessive data accumulation.
6. **Optimize Code and Data Handling:** Optimize the code to minimize unnecessary data processing and memory allocations. Use efficient data structures, avoid redundant calculations, and ensure data is properly released after usage.
7. **Increase Thread Group Ramp-Up Period:** Consider increasing the ramp-up period for the Concurrency Thread Group. This will gradually introduce more threads, reducing the load on each thread and potentially mitigating the stack overflow issue.
8. **Utilize Multiple Thread Groups:** If the issue persists, consider splitting the test plan into multiple thread groups, each handling a specific portion of the test flow. This can help distribute the workload and reduce the likelihood of stack overflow.
By carefully examining the code, data handling, and thread group configurations, you can identify the root cause of the stack overflow error and implement appropriate solutions to prevent it from recurring.
Hi Raghav, Thank you so much for wonderful JMeter videos, I got good understanding now, I have one doubt, How the same configuration will work for multiple api's in a Concurrent thread group.
Like I want to Hit my all api's with 50 parallel threads, how can i achieve this? If I run all the api's simultaneously will the load will distribute or the same load will occur on all the api's ?
Hi Arpit, In concurrency thread group, you can set the time when you want to run with full load. It is basically to mimic real world scenario, In addition you can use some controllers too
Hi i have one question?in jp@gc ultimate thread group we can able to run the 2 scripts or not and it will shows the difference in graph?
Hi Anitha, Yes, you can add diff samplers
Hi Raghav!
Really nice explanation. Could you please explain how can we achieve ramp down of thread groups in Concurrency Thread group as we did in Stepping Thread group. Since Stepping Thread group is deprecated, we should be using Concurrency Thread group right?
Hi Sajin,
When "Hold target rate" time span is reached, the Concurrency Thread Group simply stops supplying any more threads and iterating them over and over so each thread will just finish executing samplers and stop.
Hi Raghav,
Can I distribute the Concurrency Thread Group load by running it on various slave machines.. ? Please suggest...
Regards,
Santosh
Hi Shanmukha, for slave(remote testing) you can actually do from a normal thread group also. If you put 100 user load and use 2 slaves, then both will run with a user load of 100. So in total 200 users. I have not tried using Concurrency thread group for remote testing, You can try it out. Take help from this - ua-cam.com/video/Ok8Cqc0wipk/v-deo.html
Hi Raghav,
When come to real senarios i face this kind of problems.Can you please help me to resolve those.
1.I'm recording a web application using jmeter recorder.Then it will add the normal thread group.But I want to add "concurrency thread group" to there.But I want to use the "concurrency thread group".
Is that ok to add new concurrency thread group and copy all the recorded samples to there?
2.I'm using the Blazemeter chrome plugin to record because it is easy and not want to change the proxy time to time.But our office guys says ,we cant trust it because it is a third party tool.Is that true and what is the better way?
3.There is a web application in our office and I want to test the performance of a real senario.
Let's say within the given time some users are login,some are already logged and idle,some are file uploading,some are file dowmloading,some are wathing video.
Can i test this kind of senario in jmeter?
Please support me.
HI Maneka,
1. Yes, its okay,
2. In that sense, everything is third party. Even JMeter is not your in-house tool. However if there are some security concerns than it can be different, However, i don't see issues using Blazemeter.
3. Yes, you can create diff thread groups in a test plan and run them
@rahgav, thread iteration is stopped working in concurrency thread group, is there any solution for the same
Hi Sukhwinder, I will need to check on this. Pls ensure there is no setup issue at your end
Hello Raghav,
I am a beginner for Jmeter. I have watched tutorial 1-3, 6, 17 and Tips n Tricks 8. For the testing I need to run, I need login as 10 different users (with different email address and password) concurrently into a website and each user need to trigger a search. I need to check if the system can tolerate such traffic. I am unsure how I can let 10 different users to login concurrently. What tutorials do I need to watch in order to achieve that?
Thanks,
Ricky
Hi Ricky, you can use variables and refer the data from external source like excel or can also create user defined variables within JMeter and refer data. guide.blazemeter.com/hc/en-us/articles/207421395-Using-User-Defined-Variables-Using-User-Defined-Variables
Hi raghav,very amazing videos helped me alot to learn. I am stuck in a confusion. I am new to performance testing I need to know that In my jmter test plan i want to run 3 sets of users group , the test should run for 11 hours ,in which from thread 1 group every 60 sec new thread should be added , second group 2, every 2 sec thread should be added and third group also 2 threads per second , which thread type and timer I need to you ,do u have any document or reference from where I can get information
Mishka
To achieve your desired scenario in JMeter where you want to run 3 sets of user groups with different thread addition rates, you can use the following configuration:
1. Thread Group Configuration:
- Add a "Thread Group" for each set of users. You will have three Thread Groups in total.
- Configure the number of threads and the ramp-up period for each group accordingly.
2. Schedulers for Dynamic Thread Addition:
- Use the "Constant Throughput Timer" to control the rate at which threads are added. You will need to configure it for each Thread Group as per your requirement.
3. Setting up Constant Throughput Timer:
- For the first group (1 thread every 60 seconds), set the "Target Throughput" to 1 request/minute. This can be calculated as 60 seconds / 60 seconds = 1 request/minute.
- For the second group (2 threads every 2 seconds), set the "Target Throughput" to 1 request/2 seconds. This can be calculated as 2 seconds / 2 seconds = 1 request/2 seconds.
- For the third group (2 threads every 2 seconds), set the "Target Throughput" to 1 request/2 seconds as well.
Here's a more detailed step-by-step guide:
1. Thread Group 1:
- Number of Threads: Set the initial number of threads.
- Ramp-up Period: Set the initial ramp-up period.
- Add a "Constant Throughput Timer" with a Target Throughput of 1 request/minute.
2. Thread Group 2:
- Number of Threads: Set the initial number of threads.
- Ramp-up Period: Set the initial ramp-up period.
- Add a "Constant Throughput Timer" with a Target Throughput of 1 request/2 seconds.
3. Thread Group 3:
- Number of Threads: Set the initial number of threads.
- Ramp-up Period: Set the initial ramp-up period.
- Add a "Constant Throughput Timer" with a Target Throughput of 1 request/2 seconds.
With this configuration, you should be able to achieve the desired scenario where threads are added at different rates for each user group. Adjust the initial thread count and ramp-up period based on your specific requirements.
Hi Raghav,
I have a doubt, How to make settings in Concurrency thread group so that the script runs for one iteraion?
I tried in two ways:
1.By not giving anything in hold target time and by giving 1 in iteration count
2.By giving time in hold target more than a iteration time and count of iteration as 1
In First case it just replayed for 2 secs whereas in second case it considered the hold target time value and went for second iteration too.
Please help me with this ASAP
Thanks,
Rajitha
Hi Rajitha,
"Thread Iterations Limit" defines the maximum amount of loops each thread is allowed to iterate. The actual number of requests will vary depending on other settings (Target Concurrency and Hold Target Rate Time). Each Thread which will be kicked off by the Concurrency Thread Group will make not more than the number of loops you define in the "Thread Iterations Limit" section.If you are looking for "classic" iterations - it is better to go for "normal" Thread Group and Constant Throughput Timer to control the concurrency.
Ref - www.blazemeter.com/blog/advanced-load-testing-scenarios-jmeter-part-4-stepping-thread-group-and-concurrency-thread
Hi sir, I want to design an API load test (REST API) using JMeter. This is example scenario:
For the first test run, we want to run 1000 virtual users, ramp-up 30 mins, hold target rate time (sustain load 1k users) for 1 hour, and lastly ramp-down 30 mins.
How do I apply this design test? I'm pretty confused whether to use Concurrency Thread Group or Ultimate Thread Group because Concurrency Thread group don't have the Step Down feature.
Hi Anis, you can set this up using JMeter, Can check my JMeter tutorials here - automationstepbystep.com/
Thank u so much master
You are welcome Melvina
Hi Raghav,
I want to run parallel request with different data, can I do that with jmeter? Like, after login with different users, get user details api should get hit for each user respectively. All this in parallel.
Hi Himanshu, you can try parallel controller - www.blazemeter.com/blog/how-to-use-the-parallel-controller-in-jmeter
I am trying to use one of the plugin called custom thread group and trying to run it via non-gui mode, but i am getting following exception/error. Could you please help me with this? An error occurred: Error in NonGUIDriver Problem loading XML from:'/home/cloud-user/apache-jmeter-5.3/bin/LoadTestingDemo_060921_02.jmx'. Cause: CannotResolveClassException: kg.apc.jmeter.threads.UltimateThreadGroup
Hi Praveen, see if this helps
stackoverflow.com/questions/25759977/conversion-error-when-opening-jmx-file-from-jmeter-2-7-in-jmeter-2-11/25767536
@@RaghavPal ,Thanks Raghav, the issue has now been fixed.Now i am able to run it in non gui mode without any issues, i had added the plugins as suggested :) thanks for your prompt response and timely help! :)
Hi Raghav - Can you please share an example on how to use Throughput Shaping Timer along with Concurrency Thread Group to achieve better load profiles
I will plan to add on this
Hi. Thanks for the video.
I have a query about Junit integration in jmeter. Can you help me. I am facing issue in concurrent execution thread group.
Hi Rajshri, can let me know the exact issue
Hi Raghav, Great work on the video series.
I have the following use case (simplified):
1. User logs in
2. Approves a document
3. User logs out
With the ultimate test group I want to simulate an 8-hour
work day scenario where the peak will be around noon
and then ramp down until end of work day is reached.
I'm struggling as to how to define the Ramp-down period.
Will the ramp-down wait to make sure that each iteration
is completed from step 1-3 (i.e. every user is logged out) ?
Hi Jeremy, you can use Concurrency thread group
@@RaghavPal
Hi Raghav thanks allot for your response.
Concurrency Thread group has no ramp down option.
I want the target to be reached around noon and to then gradually
ramp down again, similar as to what you describe here:
ua-cam.com/video/cOPnXUDmTBY/v-deo.html
The problem is that the shutdown command that JMeter sends does not take
into account that users should log-out first. I need some kind of logout hook
that gets executed after JMeter sends the shutdown thread command.
If I dont have do logout command then my server(Siebel) gets tons of logged in users
who are actually not part of the test script anymore
Done allot of searching and I see others running into the same problem
(stackoverflow.com/questions/30242365/jmeter-execute-logout-action-during-ramp-down)
Curious to your opinion. If there's no clean solution then I'll just stick
to a linear path to the target with no ramp down using Concurrency Thread.
At the end of the day/end of the all test I will then do a cleanup of logged in users.
Looks like not a specific solution for this yet. I will check more
Hi Raghav,
I want to simulate a realistic load test where every user is having a unique login details.Testing application only accept only the registered users in the app.Therefore I could not use the CSV data set config.
Hi Maneka, this is a limitation of the test application. JMeter can handle multiple user credentials
I have one scenario I need to set up 15 minute ramp up time for 1000 Thread and then Run the test for 1 hour and after that ramp down for 15 minute. How to Configure Ramp down ?
Hi Sonu, try ultimate thread group jmeter-plugins.org/wiki/UltimateThreadGroup/
Hi Raghav,
I have one question. Is there any way we can know how many time a thread was iterated. Lets say I am running a test with 35 Target Concurrency for 1 hour. i want to know that how many time a single Thread was iterated. Is there any way of knowing that?
Hi Ashish, you can add a tabular listener or get the result in a csv file and check.
Hi, Can you please explain "Threads per request". If I'm running an API with 100 users and samples generated are 5000, then can we calculate it?
Hi, 100 users or threads means 100 users will be running the request simultaneously. Now the number of requests can by any count. It is basically the total no of request sent by the users during the duration of testing
Thanks raghav
you're welcome Keshav
Hi Raghav,
Thanks for the tutorial.
I want to discuss one scenario, where I have around 38 requests in my Thread Group. I have made the following updates in Concurrent Thread Group:
Target Concurrency - 50
Ramp up time - 5 (min)
Step count - 10
Hold Target - 1 (min)
In this scenario - the initial few requests run multiple times some times more than 50 requests. But few requests at the bottom did not execute even for single time test stops after 360 sec.
Any suggestion on how can I make to execute all requests in the provided time.
Thanks,
Shivaji
Hi Shivaji, just run it for one iteration with a single user load, without any limitations on time or no of req and check.
@@RaghavPal You mean Target Concurrency - 1 and all other parameters empty?
I have tried this but the script does not start in this case.
Hi, it was indeed a great tutorial. Do we need to install the plugins in the slave machines also. Kindly comment
will be required on slave machines too
Hi,
I require after maintaining a time the test, lower the number of threads little by little, as it could be done with a concurrency thread group
Hi Catalina, this was available in Stepping Thread Group - ua-cam.com/video/cOPnXUDmTBY/v-deo.html
Hello, thank you for this tutorial but I'm getting confused about the configurations, my target is to run 1000 user 200 user will call an API at the same time this is my configurations: Target concurrency = 1000, Ramp-up time 10 seconds, Ramp-up Steps count 5 and Thread iterations limit = 200 is this correct? thanks in advance
Hadi
Your JMeter configuration aims to achieve a target concurrency of 1000 users with 200 users calling an API simultaneously. Let's break down your configuration:
- Target Concurrency: This is the total number of concurrent users you want to simulate. You've set this to 1000.
- Ramp-Up Time: This defines how long JMeter will take to get all the target users active. With 10 seconds, JMeter will add users over this period until 1000 are active.
- Ramp-Up Steps Count: This is the number of steps JMeter will divide the ramp-up period into. With 5 steps, JMeter will add users in 5 increments over the 10-second period.
- Thread Iterations Limit: This is the number of iterations each user will perform. You've set this to 200, which means each user will execute the test plan 200 times.
However, there seems to be a misunderstanding in your configuration. If you want 200 users to call an API at the same time, you need to adjust the ramp-up time and steps to ensure that 200 users are active concurrently within your desired timeframe.
For example, if you want to ramp up to 200 concurrent users within the first few seconds and then continue adding users until you reach 1000, you might configure it like this:
- Target Concurrency: 1000
- Ramp-Up Time: 2 seconds (for the initial 200 users)
- Ramp-Up Steps Count: 1 (to add all 200 users at once)
- Thread Iterations Limit: 200
After the initial 200 users are added, you can continue to add users in increments until you reach 1000. The total test duration should be long enough to accommodate all user activity.
Remember, the actual ability to support these numbers depends on the resources of the JMeter host machine. If you're running into issues with large numbers of users, consider distributed testing with multiple JMeter instances or using cloud-based solutions¹²³.
It's also important to monitor the performance of the system under test to ensure it can handle the load and to gather accurate metrics. Adjust your configuration as needed based on the results and the behavior of the system during testing
is this useful for stress testing .? can i show this test plan as stress testing to my client..?
Yes, you can customize as per your needs
in jmeter can we execute test having test plan that has two thread group and the names of both of thread group are same??and can we execute both in same time?
Hi, you can try adding both in a test plan and run the test plan
Hi Raghav sir, this is very good training. Could advise on how to validate this load test results to determine for saying good results or bad results. Is there any % like 1 or 2% failed from overall is considered as good test results. please advise.
Hi Pavan, usually this will be decided by your team. Like what will be a good performance stats for your application. Watch this - ua-cam.com/video/f3K9gWKpY2c/v-deo.html
Thank you for details. Is there any video that helps to understand difference between loan runner vs jmeter load test. Please advise.
Not any video I created, but you should find information online
Hi raghav, hope u r doing well... The thing is when I am doing j meter concurrency threads group and am clicking on run no matter what my http request are failing....plz guide me raghav
Hi Shreetesh
Here are some possible reasons why your HTTP requests are failing when you run your JMeter concurrency thread group:
* The URL or request parameters are incorrect. Check the URL and request parameters in your JMeter script to make sure they are correct. You can use the View Results Tree listener to see the actual URL and request parameters that were sent to the server.
* The server is overloaded. If you are running your JMeter test with a high number of concurrent threads, the server may be overloaded and unable to handle the load. You can try reducing the number of concurrent threads or increasing the ramp-up period.
* There is a bug in your JMeter script. If you have recently made changes to your JMeter script, it is possible that there is a bug in the script. You can try debugging your script or running it in a simpler environment to see if you can identify the bug.
Here are some additional things you can try to troubleshoot your JMeter test:
* Enable saving response data. This will allow you to see the actual response from the server, which can help you to identify the cause of the failure.
* Use the PerfMon plugin. This plugin can provide you with more detailed information about the performance of your JMeter test, which can help you to identify the cause of the failure.
* Ask for help on the JMeter forums. The JMeter forums are a great resource for getting help with troubleshooting JMeter tests.
I hope this helps!
Hi :)
I have set Target Concurrency = 10, and Ramp Up Time (sec) = 10, I noticed Thread Group 1-1 for ex called multiple times, or Thread Group 1-2. Could you please tell me why ?
Hi Iydia, 1-1 means first iteration for user 1, 1-2 means first iteration for user 2
@@RaghavPal I know, Just i mentioned above that 1-1 called multiple times . why ?
Hi Iydia, okay, can you send me your jmx file to look into this
Hi Brother, Are you familiar on testing AWS Lambdas with JMeter?
Not yet worked on that
Hi Raghav,
I require little clarification over linking of concurrency thread group when recording is done via blazemeter.
As when we do recording via blazemeter and open that jmx file in Jmeter we already have thread group present in it. I tried to run my test by putting simple thread group as disabled and added a concurrency thread group but it gave me no results. Can you please help me with it.
Hi Neha, it creates a sample thread group with all your recorded samplers. Now I have to see your setup or replicate to check what is the real issue.
There is not any issue as script is running with 200 status. Rather my query is how can the recording done via blazemeter adds concurrency thread rather adding a simple thread group.
You can even say that i am looking for linking of blazemeter script with concurrency thread group rather than simple thread group. I want to run my script (which is recorded via blazemeter) using concurrency thread group and not simple thread group.
Okay, so that you will have to do manually in the TestPlan Neha. Blazemeter will give you the recorded test plan and you can then configure it as per your need
I tried the same but it gave me blank results in listeners. I added concurrency thread group and disabled the simple one.
Can you please help me with this
Hi Raghav!
If I want to perform a series of action on a website like hitting the url, adding items to the cart and then checkout with 100 concurrent user how can I achieve this? Please reply
Hi Mohita, you can use Include and Module controllers for this
@@RaghavPal but how to put load on these actions?
You can find some examples for these online, that will help to get started
Hi Raghav,
API is throwing 504 bad gateway error for all users when I try with 30 concurrent users. All API calls are getting passed when I try with 20 users. What will be the issue?
Hi, if this happens when you increase user load, it can be a performance issue and you can share your results and check with your teams
Nice Video Raghav. I'm new to Jmeter. I have total 50 users and in every 5 second new user should come for the login. Please suggest which thread group should i use and how to setup it. I'm getting confused in Simple thread group, Concurrency thread group and Ultimate thread group. Please suggest.
One question: In Summary report-> Sample count will be as per the request? For Example: Login page link will hit 50 times because we have 50 users. So if we are using Concurrency thread group and Ultimate thread group then login page link's Sample count is increased.
Hi Tushar, You can try with Ultimate Thread Group, the load is for user count and not request, so a user can run a request multiple times unless configured
Hi Raghav,
I need more clarity on the difference between Ramp up time and Ramp up steps.
Thanks!
Hi Rukshsar, ramp up time is the time to ramp up to complete user load. for example if you selected 100 users and ramp up time is 10 sec. So the complete user load will be done in 10 sec. That is 10 users will be added every 1 sec until 10 sec and then the test will run with full user load. Now, here without providing step count the ramp up will be distributed as per the time. However if you want the ramp up to happen in steps, like say 2 steps, Now the load will be ramped up to 100 users in 10 sec but not gradually, but in 2 steps. That is it will start with 50 users and then after 5 sec another rampup step and will add 50 more. Now here the question may be that after 5 sec the user load became full i.e. 100 users and that is correct. We have to see what we need as per our application, Usually the ramp up can be gradual.
@@RaghavPal Hi , in case for 100 users and 2 steps count, 50 users will get ramped up at the end of 5th sec , or are distributed across 5 sec.
Hi,
I need one small help from you,
I need to execute 30 concurrent requests that can be fulfilled by this video.
But I need to add the query param in my request that will be different in each request.
Let's say my get request has id as a query param.
and that will come from the previous request response.
Please help. if this is possible.
Thanks
Hi Ashwani, you can extract the value from earlier request and store it in a variable and use it in next request(s). For this you can use :
Correlation - ua-cam.com/video/7-NcyZuUQnw/v-deo.html
JSON Extractor (for JSON requests) - ua-cam.com/video/3w0x6u4-6ng/v-deo.html
How much thread groups we can pass when we are doing performance testing in local system. Is any limitation.
Hi Murali, it will depend on the system configuration, you can find the load you can put based on configuration online, I will suggest not to use your primary system for any load testing, you can create and test the requests by going max upto 10 or 50 users on your system and then for the actual load test, use another system like a remote machine or cloud systems. Your IT team can help in that
@@RaghavPal Thank you Sir
How do we simulate a Scenario where multiple users login, navigate through different pages and than logout.
Hi Nupur, can check this - ua-cam.com/video/jKiZ1efpE5w/v-deo.html
@@RaghavPal Hi, is this the correct link for simulating multiple users
How can i get interview good questions of jmetre on experience or fresher level
Is Udemy provide the interview questions.
Hi Rohit, you may find many good resources online for JMeter interview questions
If I understand correctly Target concurrency is the number of users, Right?, LET ME EXPLAIN YOU WHAT I DID............ I used for a big test, Target concurrency 1000, Ramp Up Time (sec) 600, Ramp-Up Steps count 10 (hold target and thread ite limit left empty), so I was expecting to see 1000 concurrent connections splitted in 10 steps along 600 secs (10 minutes), instead I see in my Listener "View Results in Table" that the sample column reached 44141, that is 40 times more that what I expected, definetely I'm just learning to use the tool.
Do you mind to breakdown what I did based on the numbers I inputted?, thanks.
PS: I'm figuring out that I had another normal Thread Group (under the same Test Plan) with this: Number of Thread 300, Ramp-Up period (secs)100, Loop Count 2, ........ and it seems that if I press run JMeter will launch all of them, Is my assumption correct?..
Hi Frannelk, the samples column show the number of requests made. So it can be a large number. It is basically how many requests were made during this duration. Every user can send multiple requests.
Can we check the stress testing with concurrency thread group
Hi Rohit, for doing a stress test ,check this www.blazemeter.com/blog/how-perform-stress-test-jmeter
Hi Raghav sir, how can we control the number of api requests ?
Hi Sudhir, do you mean total no of requests or Throughput
Check this thetesttherapist.com/2019/02/08/stop-jmeter-test-run-when-reaching-a-specific-number-of-requests/
Hi Raghav,
Nice explanation. I have a doubt. How do we achieve pacing in concurrency thread group? Like how do we achieve TPH?
Hi Pritam, Pls explain in detail, what exactly you are looking at
Suppose, we have requirement like #user=500, TPH= 100000. In that case, we can calculate time interval as 18 sec. To achieve 100000 TPH, we need to set fixed time interval as 18 sec in loadrunner (third option under pacing section in run time settings).
So, for example, if one whole iteration is taking 11 sec. then it will wait (18-11)sec i.e. 7 sec. In this way, we can achieve a particular TPH.
Is there any way to put the time interval in jmeter to achieve a particular TPH?
Hi Pritam, this can help you
stackoverflow.com/questions/45991690/load-distribution-of-transaction-in-jmeter
www.blazemeter.com/blog/how-use-jmeters-throughput-constant-timer
What are s distributed load and condition load on the server
We can do distributed load testing by using multiple machines and distributing the load
For having any conditions we can use controllers for if else structure
So The Concurrency Thread Group doesn’t provide the ability to define initial delay, step transition and ramp-down - which the Stepping Thread Group does means that it is not better than STG?
We are loosing some better options in this one. Don't know why they deprecated such a good thread option !
I too liked Stepping Thread group. This link can answer few queries - groups.google.com/forum/#!topic/jmeter-plugins/TF6GF2XidYQ
Getting an issue. i crated a plan with
Target Concurrency: 100
Ramp up time 40 Sec
Ram up steps 10
Hold target: 40
Iteration: 1
Action to be taken after a sample error: Continue
When execution started it never ends and keep going on. Don't know why it not stopped and why it keep running.
How add dynamic variable in concurrency thread group
For Api request
Can use functions
I use normal thread group
THREADS ARE 50
RAMP UP TIME IS 2MINUTE
LOOP COUNT IS 2
SO TOTAL NO OF SAMPLE IN THE AGGREGATED REPORT IS 100
BUT SOME TIMES IT SHOW 110
Or 105 why
Hi Rohit, have you also set the duration for the test,
How can we calculate the sample in aggregated report for concurrency user.
Hi Rohit, check this sqa.stackexchange.com/questions/37501/how-to-calculate-number-of-samples-for-concurrent-users-using-either-jpgc-ultim
@@RaghavPal thanks sir this is amazing but i want a video also
In concurrency thread group throughput is less as compare to the thread group why it is so
And how can we find the error in view result tree
will plan a session on this
Hello, can you explain the order in which threads are executed?
Hi Apeksha, I will need to check on this, Let me know if there is some specific issue with the order
@@RaghavPal if thread and ultimate thread are added in test plan along with set up and tear down.
Okay, what is the scenario to add both, you can do with one
What is the difference between the Concurrency Thread Group and Step Thread Group plugins?
Mostly do the same stuff, Stepping Thread Group is deprecated now
hello can u make video how to calculate throughput in j-meter?
I will do Abdul
@@RaghavPal thanks waiting u can make video on how to calculate it and which parametrs we should see for checking performnce of any website
I will check on thia Abdul
@@RaghavPal sure make video which important factor we should check for performnce of j meter to check web application
And the distributed mode test??
Hi Pedro, this will help ua-cam.com/video/Ok8Cqc0wipk/v-deo.html
Let me know if you have any specific questions on this
When we use this concurrency thread group ultimate thread group
Hi Rohit, I will plan for a session on this, for now this can help stackoverflow.com/questions/60411331/what-is-the-difference-between-concurrency-thread-group-and-ultimate-thread-grou
Hi sir,
I want to add 10 concurrent users and i has to run for 10 hrs continuously and iteration is 1.
Target concurrency = 10
What would be the
Ramp up time in (sec)?
Ramp up step?
Hold target rate time?
Iteration= 1
Can u please guide me
Thanks
Hi Avinash, you can try Ultimate Thread Group
@@RaghavPal thank you sir
Can i count the thread of any webserver using j meter
Hi Bullu, thread means users here that we set as per the user load
@@RaghavPal ok so i want to monitor Apache httpd servers thread.
How can i see the thread count
You can try perfmon
@@RaghavPal thanks for reply.
Can u explain it How to do.
How to test web services in JMETER 4.0
Hi Gaurav, see this - ua-cam.com/video/eaU7951fNuQ/v-deo.html
For other videos on JMeter :
JMeter Beginners - ua-cam.com/play/PLhW3qG5bs-L-zox1h3eIL7CZh5zJmci4c.html
JMeter Intermediate - ua-cam.com/play/PLhW3qG5bs-L-PpK83CC_m_bWd_722tvgk.html
JMeter Advanced - ua-cam.com/play/PLhW3qG5bs-L_Eosy1Nj1tKHC5jcBAVkPb.html
JMeter Tips n Tricks - ua-cam.com/play/PLhW3qG5bs-L-oMT1Cnrurt7gEaLrRub2V.html
All Playlists - ua-cam.com/channels/Tt7pyY-o0eltq14glaG5dg.htmlplaylists
Hii Raghav,
The number of threads users is 5000 in load testing but the run is only 2000
Hi Abhimanyu, it will depend on how you have set the load distribution in the Concurrency Thread Group, Can get some more details here medium.com/@kalyani1729mittapalli/load-testing-using-concurrency-thread-group-in-apache-jmeter-63388ac3ec17
ITS VERY BASIC THINGS.
BUT KNOW ONE SAY WITH GOOD CLARITY.
HOW TO ACHIEVE TPM ??
IN PT , IS THIS VERY DIFFICULT?.
PLZZZZ
TRY TO GIVE GOOD CLARITY
I will plan better
i have seen something , that number of concurrent users != number of requests to be done.. I am currently working on a research regarding that , i appreciate your help =) , what i want is 0-1000 thread , each does only 1 request
HI Amr, depends on your configuration. If generally define the number of users and not the number of requests, So one user can make multiple requests. You will get more information on this online.
@@RaghavPal actually i have searched alot for that, even i played with the throughput controllers and nothing works, i just cant do it with a concurrency thread group
I will check on this
@@RaghavPal I appreciate that very much
When we use this concurrency thread and why we use thread group
Hi Rohit, The concurrency thread group provides some extra options that you may not find in the default thread group. You get to select ramp up time and steps , hold target time etc
@@RaghavPal what are the extra option sir
As I said, You get to select ramp up time and steps , hold target time etc
@@RaghavPal sir what is response time and how can we calculate
Response time, you will get in the results (Listeners)
I have created 10 threads it repeated again and again and show hundred's of sample# its so confusing me can you please explain whats happening . is it connecting the target link again and again to create load ( I have created test plan for load or stress testing) these are the links of screenshot's
prntscr.com/n2hakj
prntscr.com/n2haug
Hi, this is correct, see you have set thread to 10. If you see the result, check ThreadName column, the thread count never goes beyond 10. However these 10 users working for ~30 sec can send multiple requests. (Its not that one user will send only one request and then die). Do therefore you have multiple samples. Hope this is clear
Your videos are good But reduce your pace .Its hard to understand you
Hi Deepa, you can control the speed from player settings. www.businessinsider.in/tech/how-to-speed-up-youtube-videos-as-much-as-2-times-the-normal-speed-or-slow-them-down/articleshow/70847067.cms
@@RaghavPal Hi Raghav .will try it.. Hope I didn't offend you. Do you take sessions for Jmeter and Loadrunner Then please let me know
No worries Deepa, Infact I do understand that everyone will not know of the settings and therefore in my current videos I am putting this message so everyone will know. As of now I have not planned for personal batches, You can check more info here - automationstepbystep.com/