Showing ideas with status New Idea.
Show all ideas
Currently you can only set 1 offset value for the complete test. This is not sufficient because I need to use different Offset values for different Virtual User Groups (to be able to simulate slow and fast working users) (this is because our application uses a timestamp in the users' response body to calculate how fast/slow a user is working with the application). I need to be able to set different offset values, and link this to Virtual User Groups. Somehow it would need to be possible to link Offset values to Virtual User Groups, to simulate "fast" users and "slow working" users.
... View more
Status:
New Idea
Submitted on
09-26-2017
08:30 AM
Submitted by
jose_pjoseph
on
09-26-2017
08:30 AM
I would like an option to exclude a particular page from being included in the Top 10 > Slow Pages (Average) section. Use Case: I have a test with 10 virtual users, each signing into an app and doing different things. The problem I am facing is that the sign in page (or landing page after user signs in) is one of the slow pages and after a load test, 6-8 entries in the Slow Pages (Average) list will be filled with the sign in page data from different users and the Top 10 list would look like this: Scenario1: Page 0003 (Action XXXX) Scenario3: Page 0002 (Sign in) Scenario2: Page 0002 (Sign in) Scenario5: Page 0002 (Sign in) Scenario4: Page 0002 (Sign in) Scenario9: Page 0002 (Sign in) Scenario6: Page 0002 (Sign in) Scenario8: Page 0008 (Action XXXX) Scenario10: Page 0002 (Sign in) Scenario7: Page 0002 (Sign in) Because of this, I think I am missing information related to other slow pages. I would like to either: Show "Sign in" page only once (the one that took the longest) in the list OR: After I see the above scenario, be able to exclude "Sign in" from the Top 10 list in subsequent test iterations. Also from @AlexKaras: "I think that it must be considered how to clearly indicate that the given report lists not Top 10 Slow Pages, but Top 10 Slow *Filtered* Pages, so that it is clear to report viewer that there are even more slow pages than those reported but for some reason these pages (provide names?) were filtered out."
... View more
Now when correlating data, it is possible to replace only the whole value of the parameter been correlated with the variable value.
Though it is quite a common case when only partial data replacement is needed.
At the moment, this can be solved with the help of the Set Variable Value operation (often, with the non-trivial expression).
But this workaround:
a) cannot be used in the scenarios that use Call Scenario operation; and
b) clogs up the scenario logic (especially, if more than one variable must be set).
It would be great if it is possible to use variable itself within the data correlation expression. And I beleive that this will not decrease the performance of scenario simulation.
Please see the https://community.smartbear.com/t5/LoadComplete/how-to-pass-variable-directly-in-the-Value-from-the-Request-body/td-p/174481 thread as an example that illustrates this request. It would be handy if the approach that the author of the thread tried initially is possible.
... View more
My company runs LoadComplete daily across all of our sites to check response times. Since we have over 30 websites, it's easy to see why we would like to automate this task. Via command line, we want to be able to run a test and export the results to a web page (HTML), but the only reporting options there are via command line are PDF, MHT, or XML. Although MHT is similar to what we want, most modern browsers don't know how to render them without some kind of plugin, so HTML would be the best option. I know LoadComplete can export reports in HTML in the software, so I would assume it wouldn't be too hard to add this to the command line option. If you could do this for us, that would be great.
Thank you.
... View more
See more ideas labeled with:
In the case where the result body contains a list of items returned, allow selection of multiple occurances of the ID or unique item for each row in the result body. Randomly select one of the result and return that result in a variable to be used in a later request.
... View more
This feature request was submitted on behalf of Ramyam Intelligence Lab Pvt. Ltd
Report the overall number of requests that were processed by the application server per second.
Customer's comment: This will help us tune the application's server parameters.
Thanks,
Yimy
... View more
Use case: My load tests use lots of variables. These variables often influence test circumstances, and I want to have the variables used mentioned in the Load Test Report. A good way (I think) would be to implement a new tab named Variables and create a list of used variables and their values in that section. See below for example: Note: also make this info available in the Exports (PDF, HTML)
... View more
Right now you can only set a value for hour (or parts of hour) for Offset values. For example for a 1-minute offset you have set (1/60th parts of 1) - 1 has the value of 1 hour. For more fractional parts you have define even smaller parts. Also you cannot set offset values for year/month/days Please improve this by implementing offset values for year/months/days etc in seperate input fields.
... View more
On a regular basis, Amazon images instantiated by LoadComplete crash. After test this is error message is displayed. For better analysis could the instance ID be displayed in the error message?
... View more
Use case: We run our tests on custom build Amazon Images. However, to maintain the overview, Tags are used to identify images (also used for cost / management). Please improve LoadComplete so that Tags (keys and values) can be given in LoadComplete, and these values are being set when the Amazon Image is created.
... View more
Currently, you can only check on equalness, contain or does not contain. But I would like to be able to check if a value is between specific values (eg. > 0 , < 1000). Could this be implemented in the Operations section.
... View more
In the generated Report all sort of values are calculated and displayed. For example Page Load Time, Scenario Completion Time etc. and these values are displayed at intervals of 5 seconds. But I miss overall averages (of course these can be calculated by copy/pasting the tables to Excel), but would be better if these are mentioned in the reports as well: see attached example:
... View more
We run our tests on Amazon Images. Any info about these images is lacking on the test report. In the test report, we would like to see image specifics like image type as defined in the image definition: Also IP addresses as they are generated when the test is initiated (see below): Please include all above info in the test log (for example on the Infrastructure tab (or a new tab). So a more resourceful Test Report can be generated.
... View more
We run our load tests on Amazon Images. We would like to monitor server metrics (CPU, memory, network etc) on these images and include these in the logs. Currently this is (as far as I know of), not supported by LoadComplete. Please include this feature in upcoming version of LoadComplete.
... View more
I would like LoadComplete's Loop function to be improved so that it's value can be randomized (determined at runtime), between a set min and max value. The reason for that is that I want my user(s) to perform recurring tasks (thats the reason why I use Loop function) in different loads (for example: simulated user 1 performs the tasks in the loop 4 times, simulated user 2 performs these same tasks 8 times, simulated user 3 14 times, etc. I see LoadComplete enhanced so that min and max values can be set, and a checkbox that can be enabled/disabled to calculate random values (at runtime) and executes the tests (the loop) with these pre-testrun calculated loop values.
... View more
Currently it is not possible to combine random Page times with fixed think times. For example, I have following scenario (see attached screenshot): I have 2 Pages, I want the first page to have fixed think time of 6000ms, I want the second page to have a random think time between 10 and 500ms. Note: There might also be created an option to switch between fixed/ random times on the Test Editor:
... View more
In the Details section of the test log, you can export to CSV and XML. However, these functions do not export Duration, ThinkTime, and Response Time columns and values. Could LoadComplete please be upgraded so that these values are included in the export (for both CSV and XML files exports) ? Thanks.
... View more
This feature request is based on following: I need to implement a validation, based on value of ContentLength. The test should fail if ContentLength equals 0. I have read the info about validation here. But this only works for validation on strings. I want to have a validation based on numeric value of a response. Of course, the value of the Contenlength changes for every request (it's not a fixed value). Only thing that counts is that validation should pass if ContentLength value > 0, and that it should fail in all other cases. There is currently no easy way to implement such validation. I would like LoadComplete to be improved with such easy validation. See attached screenshot for clarification.
... View more
Hi,
It would be faster and more convenient if it is possible to select more than counter (using regular Shift- and Ctrl-clicks) in the Select Performance Counter window when adding New Performance Counter to the project.
... View more
We have project and scenario variables. Why not test variables so we can set things specific to the test while using the same scenario. In my example, I want to set the zipcode at the test level, since each test is run in a different location and the search criteria includes zipcode.
... View more