Is this happening every time you load/run your scripts? I get these after a while working with TC (especially if the scripts are J/C#/C script). I use a memory management program to help free up the memory not used by it. I also use my G15 keyboards LCD display to display how much memory TC is using so I can preemptive free memory. Sometimes though it will crash no matter what though. Just save/backup a lot and free up memory when you can. ('save/backup a lot' refers to when you get these errors during saving and your scripts magically disappear)
Let me know in case of following what does TestComplete do ?
// Following are declaration for arrays.
var fingerBiometricExpected = new Array();
var failedItems = ODT["Classes"]["NewArray"]();
/// some code here
Will TestComplete release the allocated memory for arrays while exiting a function or it is to be released explicitly.
Since TestComplete uses the Microsoft JScript engine, memory is managed by the script engine, and it is supposed to be released automatically by the garbage collector. I recommend that you see the following newsgroup thread for information which may help you avoid the "out of memory problems": http://www.automatedqa.com/forums/d.cgi?...34082&utag=
As for the other problems, most probably, they are not related to each other.
[ QUOTE ]
1) Access violation at address 00000029.Read of address 00000029.
[/ QUOTE ]
I guess, the error message is displayed by TestComplete. If so, to help us investigate the problem, please follow the steps below:
[/ QUOTE ]
There was a similar problem in earlier versions of TestComplete 7, and we have fixed it. Could you please check whether the problem persists in the latest version of TestComplete 7 (v. 7.20)?
I actually had this same issue and was directed by the support team that it was a problem with name mapping. Are you using name mapping?
Usually it's something that's under name mapping, that is not under aliases.
I used aq time to find the exact error and had sent it to the AQA team. After their response I was able to find the different between name mapping and aliases and fix the problem, and never got the index out of bounds again.
Like you stated it was very sporatic for me, but ended up not being a real memory issue i guess.