At 18:44 2017-03-14, Kurt at VR-FX <vrfx(a)optonline.net> wrote:
>OK - since you DID ask. FYI - I actually Step thru the code - and
>when I do - everything works fine! That's why its an even stranger conundrum!!!
>
>Also - as previously mentioned - I actually Was testing on a machine
>that has the problem. And, even on there - when stepping thru the
>code - the problem then does NOT occur...
>
>:-(
This has the smell of a race condition. You might not be able
to get at the actual cause. You may have to -- shudder --
kludge. If someone comes up with a non-kludge, use that in preference to this.
On a misbehaving system, try adding a sleep(10000) between each
statement. Does the problem still occur?
If so, come up with something else to stick in a delay that
gets rid of the problem. If it is response-related, try a
keyboard-messagebox() sequence.
If not or after finding an alternative delay method, start
removing the delays until you find where the problem triggers. Keep
that one. You might shave the time down some, too. (I selected 10
seconds as overkill*.)
After getting it to the minimal number of delays, try removing
them all. This might work. Why? Who knows?
* Gene's Law of Overkill: First, you have to kill. Only then can you
overdo it.
[snip]
Sincerely,
Gene Wirchenko