OK, I just proved a real problem.
I commented out the next couple lines, including the dim statements I showed earlier. Now the debugger still stops on the commented line! Worse, the yellow doesn't match the current line...it is shorter and indented, so I have commented code extending from both sides.



>Hold on, it gets worse!
>OK, I got to thinking that the problem may be with the file i/o, so I commented out the line and replaced it with this:
>
>

>dim g_len as integer = 200
>

>
>Pretty simple. It still stops the system in the same manner! I think it's the spot in the code, not the actual code that I see. I go back to the fact that the CLR or something in the background must be corrupt. Is there any validation that can be done to the metadata to ensure that it compiles correctly?
>
>This is turning into a huge time sink.
>
>
>>I just restarted my computer and got the same thing (the server and my test environment are on the same machine).
>>
>>Here's a bit more info that may or may not help:
>>
>>It appears my problem may be with a file operation I'm attempting. Here's the code:
>>
>>

>>dim gnfilehandle  as system.io.filestream
>>gnfilehandle = fopen(str_savedfile)   'this works fine at this point...file opens normally
>>dim g_len as integer = gnfilehandle.length -1  'this is the line that stops
>>

>>
>>Once I hit the last line there, the system sits for about 15 seconds, then my watch box shows that the gnfilehandle.length now shows "error:cannot obtain value." Between the fopen statement and the bottom line, the watch showed the correct length. My file "disappears"
>>
>>the file is in the Windows/temp directory.
>>
>>Remember that yesterday this all worked correctly.
>>
>>ugh.