Ok, we have a few crossed wires here, so let me explain. We had a problem with the 4D clients locking up so Alex and I, I was remote from this machine, did I few tests and found that 4D was freezing. About the same time I found a bug in my code that was only appearing on the live machine where the 4D client appeared to freeze but was really running a very nasty query. To compound the problem the user would try and run this nasty query 4 or 5 times before giving up. I fixed this bug and assumed it was the same as Alex's freezing, since it had the same symptoms. (4D using up all of the CPU time) The difference in my bug was that the machine would still processing, albeit very slowly. In Alex's case 4D would no longer respond at all.
The problem that I was talking about in this thread was that 4D was randomly crashing, in that unique disappearing act that 4D on Windows likes to do. I hope this clears things up. Michael ----- Original Message ----- From: "Aparajita Fishman" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, December 11, 2002 7:59 PM Subject: Re: [Active4d-dev] get global with index > > The data that was guarenteed to crash Active4D will be very likely to > > contain characters with accents, or other non printable low ASCII value > > characters. > > I would say this has a .001% chance of having anything to do with it. > At no point in my code do I touch the contents of an array element, > other than to move it as a block, in which case it is being treated as > raw bytes. > > Also, Alexander said that 4D was not crashing, it was freezing. There > is a big difference! On examining my internal semaphore management I > saw immediately why it would freeze, and I am working on a fix.
