Qlik Sense Scalability Tools

[Daniel_Larsson]

This package (referred to as Qlik Sense Scalability Tools) contains a complete set of tools for easy creation, execution and analysis of load/performance tests.

This tool is now deprecated and will not receive any further updates, please use the Qlik Sense Enterprise Scalability Tools instead.

Supported versions of Qlik Sense: all 2020, all 2021, 2022-aug

Included parts are:

• Standalone application for creating and executing a simulation script
• Documentation on how to use the package
• Regression analyzer
• Benchmarking package
• App evaluator package

QlikView and Qlik Sense documents to help analyze result and log files (previously included in this package) can be found here :https://community.qlik.com/docs/DOC-15451

Troubleshooting

For help to troubleshoot connection problems, please review Appendix A of the documentation or Connection Troubleshooting Tips

Change log

v5.17.0

• Add support for Qlik Sense May 2022 release
• Add support for Qlik Sense Aug 2022 release
  

v5.16.0

• Add support for Qlik Sense Feb 2022 release
  

v5.15.0

•  Add support for Qlik Sense Nov 2021 release

v5.14.0

• Add support for Qlik Sense Aug 2021 release

(See Readme.txt for changes in earlier versions of the tool.)

Your use of Qlik Sense Scalability Tool will be subject to the same license agreement between you and Qlik for your Qlik Sense License. Qlik does not provide maintenance and support services for the Qlik Sense Scalability Tool, however please check QlikCommunity for additional information on use of these products.

[Patrick1210]

Thanks a lot  @Daniel_Larsson 
I will keep an eye on ressource, although I am planning low scenario. I will install on central node as well.

So if I can not use ticket auth, and not using header for security reason, how can I set up the NTLM connection ?
Should I create a specific vritual proxy for NTLM usage ? how can I set up it ? I can see that TLS 1.0 should be installed according to the documetnation, and I am not sure what it is.

May I ask if I should consider to not use the scalability tool related to the environment I described. Should I consider an other option for warming ?

Thanks again for your precious help.

[Daniel_Larsson]

sorry might have confused things a little bit. So we don't support any form of "ticket" other than then what can be done by the OS itself through NTLM. So multiple users cannot be simulated etc. Since this is not the case here, you can run with ntlm provided the tool is run by a user which have access to sense. Most likely your "root" proxy is already setup this way, if so, just change connectiontype to NTLM and leave VirtualProxyPrefix blank (unless you specifically want to use a virtualproxy for other reasons).

The TLS 1.0 is no longer correct, we only support TLS 1.1 and 1.2 nowadays, I'll make sure documentation is updated with this for next release. Most likely your server has support for this already.

[Daniel_Larsson]

Both QSST and QSEST are well suited for being used as cache warming, once you have a working scenario, the normal thing is to schedule a run with the underlying command line tools (SDKExerciser for QSST, can be found i subfolder structure of the tool and gopherciser for QSEST).

As a separate trouble shooting tips, if the "connect" function doesn't give you a good enough error, you can sometimes get an better error message by running a scenario with only a short "TimerDelay" in it, nothing else. This would basically test the connection and possible give you a more useful error.

[Patrick1210]

Many many Thanks @Daniel_Larsson 

So I follow your advice on NTLM connection, without prefix on central node. but I am still not able to get through.

actually I thought I can try QSST on my own laptop which can open the hub and application with my user account. I tried to connect to the app within the NTLM option in QSST. But still not able to connect (see picture below).

I dont'get where I am wrong. As you can see on the picture, the bottom one is with the proxy ticket while the top one is withtout the ticket proxy.

in both cases, I am not able to connect.
Many thanks for your help

[Daniel_Larsson]

what error message do you get creating a scenario which only contains a timerdelay using those settings.
i.e. instead of pressing the "connect" button, instead press the "start executor" -> execute.

[Patrick1210]

Hi @Daniel_Larsson 

thanks to your advice, I made it !
I was able to connect with NTLM connection, an d to connect to an app in the everyonestream

Please, let me know if I understand correctly your great advice.
NTLM connection is using the user account connected to the server (in my case the central node). so the account I used should be able to access to all streams if I would like to open and cache any application. A service account with ContentAdmin role is ok ?

Then I will create as many as windows batches as needed and schedule them as per this post with Task Scheduler. Is it correct that the NTLM connection will keep the username in the json file but not the password. So, when creating the batch job, no need to embed any credentials to execute the job.

Final questions (thanks thanks thanks again for your answers), as I could not find so much things on caching apps, I am not sure about my settings of scenario
I would like 1 user to open an app, and then open 3 sheets of this app. Indeed, the app is really long, so I would like the scenario to wait for 1 min to ge the app open, and then 1min for each sheet.
I don't get how to fix the ramupDelay. I chose 1 instanceNumber to be run during 300s with 1 Iterations each, 1 concurrent User. is it a correct set up ?

And how can I ensure that the scenario would have finally well opened the sheets after execution ? I am not sure to find this in the generated log.

Many Thanks for your time, you are a precious help !

[Daniel_Larsson]

When using the tool in NTLM mode it will use the user which started the tool, and the user defined in the script will not be used, which means no need for embeded credentials.

When running in NTLM mode the user running the tool needs to have access to all apps being accessed.

If running from task scheduler, make sure it's run by a user (with app access) and not by "system".

The tool will not continue to a subsequent action until the current one is considered finished or timeout is reached, since tool is normally used for performance testing, this means all objects etc fully loaded. Hence an open app followed by 3 changesheet's will load the app into memory and and all supported objects on the sheets.

rampupDelay is a delay time between users being simulated, with 1 user this do not come into effect.

instanceNumber is to make sure diffrerent randomizations is done when running more than one instance of the tool. So any number here is ok for your case.

1 concurrent user is also what you want for this, so that's correct.

[Patrick1210]

Thanks again @Daniel_Larsson 
thanks to you, I have been able to create my scenario, execute and plan a batch. many thanks!

last question (please):
I created a scenario to open few sheets in one application. The logs do not mention any errors after execution of the job. 
However, when checking in the application Operations Monitor, I can not see that the application  & the selected sheets I pre-cached have been opened .
is it normal ? Or how can I check the sheets of the application has been well opened ?

Thanks again,
Patrick

[kss7374]

Hello guys!

I have a problem when testing an uncached application using QSST (Global Search).
The script fails with an error - timeout limit when testing intelligent search (GlobalSearch)
The test is performed for one iteration and for one user. A timeout of 5-10 seconds is set between each step.
When the script is run again (cached data), it works fine, without an error.

Tell me how you can change and increase this waiting parameter (15000 m s), and how to fix the duration of execution (GlobalSearch) for uncached data, so that you can then compare it with the duration of cached data in the Scalability Results Analyzer application, and see the difference?

2022-02-13 18:00:02.6039+03:00 44147102727625 SCENARIO 215 1-1 5c1d14fa-2959-424c-a701-de5e76d73284 1_MyApp ClearAll #7 ***7_2_ClearAllFilter #7 8 447703 6699 End 65
2022-02-13 18:00:02.6039+03:00 44147102727726 SCENARIO 1-1 5c1d14fa-2959-424c-a701-de5e76d73284 1_MyApp TimerDelayItem - Constant(10) #8 ###Delay10s #8 9 Start
2022-02-13 18:00:12.6197+03:00 44147128778153 SCENARIO 10000 1-1 5c1d14fa-2959-424c-a701-de5e76d73284 1_MyApp TimerDelayItem - Constant(10) #8 ###Delay10s #8 9 End

2022-02-13 18:00:12.6197+03:00 44147128803054 SCENARIO 1-1 5c1d14fa-2959-424c-a701-de5e76d73284 1_MyApp GlobalSearch #9 ===7_3_IntSearchYear #9 10 Start

2022-02-13 18:03:04.6255+03:00 44147576739308 SCENARIOERROR 1-1 5c1d14fa-2959-424c-a701-de5e76d73284 1_MyApp ERROR Search timeout [15000]:Search timeout(Search query took too long) Qlik.Engine.MethodInvocationException: Search timeout [15000]:Search timeout(Search query took too long)\n at SDKExerciserService.Driver.Workers.ScenarioItems.SessionState.<WaitForTasks>d__106.MoveNext()\n--- End of stack trace from previous location where exception was thrown ---\n at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()\n

....

SDKExerciserService.Driver.Workers.ScenarioItems.Scenario.<Execute>d__15.MoveNext()\n--- End of stack trace from previous location where exception was thrown ---\n at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()\n at SDKExerciserService.Driver.Workers.ScenarioWorker.<Execute>d__7.MoveNext()
2022-02-13 18:13:04.6426+03:00 44149139306231 SCENARIO 1 ErrorCount
2022-02-13 18:13:04.6426+03:00 44149139306287 SCENARIO 19 WarningCount

Thanks,

Sergey

[Daniel_Larsson]

---

However, when checking in the application Operations Monitor, I can not see that the application  & the selected sheets I pre-cached have been opened .
is it normal ? Or how can I check the sheets of the application has been well opened ?

I don't know much about the operations monitor, but if you can see things there when a user does the actions, you should also be able to see it when done by the tool. Make sure the app has been reloaded after the actions though of course.