> The source is a VFP cursor created by querying an MSSQL db. My app reads
> from & parses the data in the cursor into native VFP tables stored
> locally. There's no user interface. Crash about every 15min.
If it's that re-creatable, then it's time to start adding trace code,
IMO-- to find where the problem is happening. I use STRTOFILE for
this:
STRTOFILE("Some meaningful string" + CHR(13) + CHR(10), "c:\out.log", .T.)
This will append a line to the c:\out.log file-- change your strings
to something useful to know where in the code it's happening(put it in
key spots right before something you suspect to be the potential
problem-- it can be very high level in the code, especially at first).
You can also write a more robust tracing mechanism, which I plan to do
for my next projects. Basically, you call the function throughout your
code, and the function queries a setting to see if tracing is 'on',
and potentially at what level, and if so outputs it to the text file
with a date/time stamp, etc. I've used it in C++ programs as a
debugging mechanism and it's been extremely valuable. I turn it on
with a registry setting. One of the key things I like to do is add
start:functionname() and end:functionname() things to the beginning
and ending of functions/methods when I write my code initially-- this
lets me know very early on what overall function/method the crash is
occuring in...
--
Derek
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/profox
OT-free version of this list: http://leafe.com/mailman/listinfo/profoxtech
** All postings, unless explicitly stated otherwise, are the opinions of the
author, and do not constitute legal or medical advice. This statement is added
to the messages for those lawyers who are too stupid to see the obvious.
