Suggestions for future of Load Complete

Hi,

This thread is not to have any questions answered, but simply to give an opinion on the state of Load Complete at this point. We have been using it at our business for exactly one year now. We had originally chosen Load Complete over Visual Studio 2010 for Load Tests as we had evaluated that this was a better choice for our needs. One year later, I am now re-evaluating our decision as there are many pain points with Load Complete, that make running load tests quite the trouble some times. 



Here is a list of suggestions for the next versions of Load Complete and I encourage others to reply to this with their own so that the people from Smartbear can see them. 

1. 
   Connecting to servers with different user credentials: When adding counters to monitor while load testing, there is no way to add credentials to connet to those servers, meaning the user which is running Load Complete has to have the rights as admin or as a performance monitor on the remote server. Currently the only work around is to do a "shift + right-click" when starting Load Complete and running the application as the user which has the rights on that server. This is a pain to do plus it does not help us in case we want to monitor different servers with different credentials at the same time. This is one of our biggest pain points, as we have servers on many different domains with different user accounts that must be used. 


2. 
   When starting Load Complete as another user, some of the options cause us problems as we are not the local user. Exemple: exporting log results to pdf is no longer available and thus we have to use the .mht exportation which is a whole other problem in itself (format that only works with internet explorer). 


3. 
   Error messages such as connection errors in Load Complete when trying to connect to remote stations or servers are of little or no help when connection fails. Exemple: If you add a station to the Stations page and run the "check current station" you either get a red "X" or a green checkmark. When you get the red "X" you don't know why. Is the stationg offline, is the station not running the Load Complete agent, is the station not running the same version of the Load Complete agent (not compatible between versions), is the Load Complete agent crashed (this happens way too often), etc. We can loose a lot of time figuring out something that should be trivial.


4. 
   When connecting performance counters, the information given is all but useless when the operation fails. I still have some servers which I cannot monitor through Load Complete and even with the help from tech support we do not know why. I must open the Windows Performance Monitor each time I run a loat test against this server and connect to it that way. 


5. 
   There is currently no way to connect to a Linux server for performance counters. I know the windows performance couters interface does not exist in Linux, but it would be nice to have some way to connect to these, either by installing a remote app on those servers that can be connected in Load Complete or by giving the appropriate credentials to Load Complete so that the app can connect and retrieve the appropriate information. 


6. 
   After a Load test when we see the error messages (OK, Warning or error) it would be nice to be able to sort the messages other than just but status. Running a stress test creates thousands of messages and it would be nice if we could sort by return code, virtual user group and scenario. 

This is everything I can think of off the top of my head, but like I said, I encourage others to add to this so that SmartBear can add these features to Load Complete. 



Thanks, 

Louis-Philippe