Stephen,
Is there some way you can use SQL passthrough to do queries/updates? The
SQLEXEC function provides some potentially useful returns: -1 for connection
level errors (AERROR retrieves), 0 for still executing, or the number of
rows affected if successful.
Sam
----- Original Message -----
From: "Wolfe, Stephen S GS-11 6 MDSS/SGSI" <Stephen.Wolfe (at) macdill .DOT af.mil>
To: "ProFox Email List" <profox@leafe.com>
Sent: Tuesday, February 28, 2006 8:03 AM
Subject: RE: VFP8: SQL Server connection (dsnless) made through a DBC
question
Good post, Ted, I have to go see my doc (bad earache, I was out
yesterday) and I'll pick this up tomorrow. And, yes, I'm a small cog in
a huge machine. It is already been determined that the bandwidth is
swamped and they are in the process of upgrading to a 10MB connection.
I still need to do something for the users while DISA takes it good ol'
time in making the connection upgrade (approximately 3 - 5 months).
However, I'll try to provide you with a little more detail:
1. I don't use a central data manager, although I do use an application
object to open the DBC and develop the various RVs.
2. I'm not catching errors when I do a requery on critical views
(patients in waiting room type of views). So, I'll need to do that.
V/R
Stephen S. Wolfe, GS-11, DAF
Data Services Manager
6th Medical Group
Comm (813) 827-9994 DSN 651-9994
-----Original Message-----
From: profox-bounces (at) leafe .DOT com [mailto:profox-bounces@leafe.com] On
Behalf Of Ted Roche
Sent: Tuesday, February 28, 2006 8:43 AM
To: profox (at) leafe .DOT com
Subject: Re: VFP8: SQL Server connection (dsnless) made through a DBC
question
On 2/28/06, Wolfe, Stephen S GS-11 6 MDSS/SGSI
<Stephen.Wolfe@macdill.af.mil> wrote:
>
> Without doing massive redesign (I use a DBC and remote views, they
work
> good for me) how can I catch this disconnect and reestablish a
> connection with an appropriate user friendly message (feedback) to the
> user.
>
You're trying to solve a low-level problem (network drop-outs) in a
high-level language.
Ideally, you could get logs from a machine at the remote site and
document to the network administrators that they are getting ripped
off on their T-1. As Whil indicated in another thread, seven seconds
downtime in three years is more like the performance you are paying
for. There is a good chance that there is defective equipment (or
possibly network configuration) somewhere between there and there that
ought to be fixed. It's more likely to get worse than better.
If that's out of the range of possibilities, and you are a small cog
in a vast machine (I know the feeling), in your error handler, catch
these errors, pop up a huge form stating "LOSS OF NETWORK
CONNECTIVITY... RECONNECTING" and restart your application.
Okay, more practically, what is the error? Are you using a data
manager,so there's a central place to detect the error?
--
Ted Roche
Ted Roche & Associates, LLC
http://www.tedroche.com
[excessive quoting removed by server]
©2006 Sam Thornton |
