New Idea
0 Kudos

When editing a tests requests, allow the users to set up the screens as they need them for that session or user, and have a lock button so that for each request, the work area remains in the same configuration, so that they don't need to keep changing things around  each time.


A found in one session where I had to check every request I sent 20% of the time changing the environment around.

Report comparison handling

Status: Implemented
by Staff on ‎03-13-2015 01:09 PM

In the Compare Results interface, it would be nice if users could easily specify which report will be Report A and which will be Report B. As it stands, whichever report is selected first becomes Report A, and it's confusing to figure out how to rearrange test results.

Status: Implemented

Implemented in LoadComplete 3.2

0 Kudos

This would allow the test to be edited with any common mass update facilities, without having to include them in LC.

The import facility could then perform a validation to ensure that the test was compatible with the proprietary rules in LC and the tests could be used in their updated states.


The validation routine could then add stamps to indicate:

1. It has been edited with ... <tool name>, date time stamp and reference the original test.

2. A Label to allow brief notes on what was done

3. A flag to indicate it has been externally edited, 


Not all required, but these would help in mixed test environments.


Alternatively: Make tests editable with other editors directly (probably more difficult, and with less control of content).

0 Kudos



Initial clauses:
Consider this use case as a sample: login - get one page with items list - order one item - purchase ordered item - logout. The root idea is that because of reusability and maintenance considerations, it is better to record and execute not a lot of similar complex end-to-end scenarios, but several smaller separate stand-alone ones and combine them using the Call Scenario operation.
It is expected, that using this approach it will be possible to create, for example, these additional scenarios:
-- login - logout;
-- login - get random number of pages with items list - logout;
-- login - order random number of items - purchase items - logout.
From the reusability point of view it is expected that if, for example, the login scenario changes, only this simple scenario has to be recorded and correlated a-new but not all affected large and complex scenarios. It should be noted here that while it is possible to refresh the correlation rules for the already existing scenario, this does not apply to scenarios launched from the Call Scenario operation.

The above considerations are, probably, the major cornerstone for my feature request, so please let me know if my initial idea is wrong, not recommended or is not adopted by real practitioners because of some reason.

If the above idea is an acceptable one, it is obvious that data correlation will be required when combining separate scenarios to work together. I believe that this can be done more or less easily with the existing functionality of LoadComplete when complete scenario is recorded initially and is split later to several smaller stand-alone ones.

Now the problem:
Assume that in the course of the project, the page that displays items list (Items List page) was modified and requires different set of requests to be displayed. This means that the scenario that obtains this page must be re-recorded and correlated with the existing scenarios.
In its turn this means that (some) of the Data Selectors and Data Replacers that existed in the initial version of the Items List page must be recreated for the updated version of the page.

Actual request:
In order to make it more easy and convenient to find Data Selectors and Data Replacers in the initial version of the page, create them (only the needed ones, probably, with some modifications) in the new version of the page and apply at proper places, some functionality for side-by-side comparison of the requests that belong to two different pages with the possibility to modify requests of the target page should be useful and helpful.
Could you please consider this?

Thank you for your consideration.


P.S. This feature request is based on the Case #00222701 and might be related to the thread.


Dear all,


I am unable to record scenario for media streaming. Please assist me


For more information download the video :


As suggest by Alexey Kryuchkov  submit a feature request to the following Community.

0 Kudos

In order to be able to create more real-life test scenarios I would like to ask to make it possible to use variables as operation parameters.
For example, a sample scenario is to make a user to order a random number of items.
The idea of how to implement this is to record the ordering of one item, add the Loop operation to the scenario and move relevant requests into the loop.
The current problem is that the loop can repeat only the fixed number of iterations.
Expected improvement is that it will be possible to create a project variable of random type in the range, say, from 1 to 10 and reference this variable as a loop counter.
The expected result is that every virtual user will order random number of items in the range between 1 and 10.


0 Kudos

When setting up a run of tests allow the user to apply a filter to the output to the log file allowing them to filter on

1. Duration

2. OK/Warning/Error

3. Ranges of Request ID/ Time period/ Call Types/ Machine/ Test/ and so forth


This will reduce unwanted data, allow tests to be for longer before exceeding log file size limitations.

In addition, when the data is exported, the data will be ready for processing and require less time to use.

Please add an "Errors" section to LoadComplete reports.  Currently, there is only a pass/fail section.   This tells you how many requests passed and how many failed - but does give a breakdown of the errors showing why they failed.  It is important to know if the app was experiencing server error 500's or merely redirects.


Things to include:


Errors per page chart: 

Show each page and how many errors it generated


Pie chart of error types: 

Showing the percentage of each error type.


Error Detail:

Show each error.  Include request header, response header, response body (snippet) - this can be limited to the first nn errors.

0 Kudos

Please implement support for manully resetting web proxy, just in case loadComplete is broken in the middle of recording.

0 Kudos

I would like to ask for the possibility to concatenate the values of two variables into the third one. This can be implemented in the same tables where data are extracted from the responses and assigned to the variables (Data Selectors, etc.).

In this case, the value of the 'Source' field (using Data Selector table as an example) may be 'Variables', 'Regular Expression' will specify the name of the first variable (e.g. @Var1), 'Subexpression' will specify the name of the second variable (e.g. @Var2) and the 'Save to Variable' field will specify the name of the variable that contains concatenation result (e.g. Var3).

It would be perfect, if it is possible for Var1 and Var2 specify not only variable name, but some string that contains (or not) the variable - e.g.: if Var1 = 'CaseId' and Var2 = '123', then '@Var1' and '=@Var2' should concatenate to 'CaseId=123'.


Please let me know if you need more details for this request.

Status: Implemented
0 Kudos



Please guide me if I missed this functionality in LoadComplete, but I think that variables created using the Variables panel are lacking some dynamics.

For example, I often need the random date that is not in the future but may be equal to today. The current implementation of Variables makes it possible to generate a random date variable, but only within hardcoded dates range.

I think that it might help if it is possible to implement some set of simple macros in LoadComplete so that, for my example, it is possible to specify, say, {CurrentDate} in the To field of the Variable Wizard and get a random date value from some date in the past and up to today.





I would like to ask for the possibility to copy and paste functionality to be able to copy requests to another page and scenario.

There is a Copy command in the context menu for request in the Scenario Editor, but the Paste menu command is always disabled.

(Not sure if I remember it correctly, but it seems to me that this was possible in the early LoadComplete 2.x versions.)



0 Kudos

One feature I use regularly in other tools is the ability for regular expressions to capture a range of values. For example, if a web page had several menus and sub menus as a FAQ, I would use a regular expression to extract all links on a page, then click one randomly, then do the same on the sub menu and so on. 


This could be done by storing every link in a datatable and using a random selection, but each time the pages were updated the datatable would have to be updated as well. Using random responses from a regular expression is much quicker.


With Loadcomplete only allowing the choice of the first pattern match this approach is not possible.

When simulating 2 virtual user groups in parallel, it executes segment by segment sequentially. But its better to execute simultaneously to analyze the performance.



Please implement support for testing applications that use client-side certificates

0 Kudos

Step through Debug option

Status: New Idea
by ztilleto on ‎04-29-2015 06:32 AM

There should be a single step debug option where you at time of post and respons can see what was sent and returned, and then see what was available at the time the variables was to be set.


I think that would be a welcome addition as most of your other tools also allow for a step through debugging for verification of the testplan.




0 Kudos

Perform Load test on SAP

Status: New Idea
by Daoulas on ‎09-18-2015 03:04 AM

Perform Load test on SAP , which is a client application and does not run from a web browser.



When the scenario recording is made for the dynamic applications with a lot of async Ajax requests, the recorded scenario includes a lot of these supplimentary requests that, I think, may be excluded from the final test. The problem here is that it is not always evident whether or not the given request or page can be safely excluded from the test.

My current approach is to Save As... current scenario as a backup, remove requests/pages from the original one and verify if the obtained result is what I was needed.

It will really help if it is possible to enable / disable certain requests / pages within scenario like this is possible for the host server.


Thank you for considering this request.

Status: Implemented

Adjust scales in Custom Charts

Status: Implemented
by Staff on ‎08-19-2015 07:22 AM

For some metrics like %CPU and %Disk Time, the scales created in Custom Charts do not make intuitive sense. In the below example, the scale for %CPU goes from 0 - 600. This makes it harder to visually interpret the chart.


Would be nice to fix this for metrics that are percentage values, or perhaps give the user the ability to manually adjust the scales. (This could also be useful to create a unified report in some fringe use cases of running multiple simultaneous load tests against the same server.)


Status: Implemented

Implemented in LoadComplete 3.2