>>>>>>Did you test your PDF2PS function standalone?
>>>>>
>>>>>
>>>>>Yes, If I am not using Parallel Fox the same is running Fine. It seems to me that when there is One CPU involved, the DLL is not shared and may be when there are more than One CPU there may be Sharing Problem of DLL , so error might be generated. Is it so ?
>>>
>>>If you're sure that this code works fine if not running parallel, one of the possibilities is to wrap the code to call this function in try/catch block and retry in case of the failure the second time.
>>
>>I'm not sure this is good advice. I've had a problem with bullzip where the output of two processes was mixed. I had to use a mutex.
>>
>>Nobody knows what the consequences are when exceptions are being swallowed. You may not see a problem but that does not mean there isn't any
>>
>>Seems to me that this whole story of manipulating postscript files is turning out to be a joke - it smells all over from the beginning to the end
>>
>>I've posted a couple of messages in the past (message#1580759 and message#1580549) but refrained from posting afterwards
>>
>>http://idioms.thefreedictionary.com/You+can+lead+a+horse+to+water%2c+but+you+can't+make+it+drink
>
>I agree, but I do remember in the past that that particular problem was solved with retry, thus I suggested it. There is a possibility it may help.


You do not know what happened before the exception. It may have allocated resources (memory, ...) that won't be properly released. The exception will probable be the famous C0000005.
By systematically ignoring the exception, you may get in an unstable situation - without knowing this is

So it only appears to have helped


Well, it's not my problem. And I am glad it isn't