[plug] web page problem

[Leon Brooks]

russ wrote:
> On Thu, 06 Jan 2000, Leon Brooks wrote:
>> It means (1) that the person who wrote that ASP doesn't understand
>> locking and/or (2) the ODBC layer they used didn't understand locking
>> (ie didn't do a BEGIN WORK ... COMMIT block around their update) or made
>> a mistake or (3) MS SQL is stuffed.

>> It didn't happen with Netscrape simply because it doesn't happen every
>> time. If you tried again with KFM, it would be unlikely to recur (unless
>> the ASP writer stuffed up _really_ badly, which is not beyond the realms
>> of probability. (-:

> It happened 4 or 5 times in a row. I then opened Netscape
> and it worked. I closed Netscape and tried again with KFM
> and it happened again. That's when I thought I'd ask.

Hokay, in that case the ASP code in question has special cases for
special browsers and either the default case (if there is one!) doesn't
work, or they have a special case for KFM (unlikely) and that is the bit
that doesn't work. Mind you, it's unusual to find a special case for
Lynx, either. It probably comes down to something in the Browser ID
string; maybe the ASP fumbles that and decides that it's dealing with
"Netscape version 1.167.2.21" and _then_ screams and sticks it's head
into a bush (I know I would!). It wouldn't surprise me to find that the
ASP code was deadlocking _itself_ or really trying to access a
nonexistent record or suchlike, thereby breaking either ODBC or MS-SQL.

-- 
Confidence is the feeling you have before you understand the situation.
If at first you don't succeed, try a shorter bungee. When in trouble,
when in doubt, run in circles, scream and shout. The two great secrets
of success are: don't tell anyone everything that you know.