FreeTDS Development Group

[Dan Mohn <mohnd AT egolfscore.com>]

>> When you connect DSN-less, you're relying on the built-in TDS
version. Your package might have used the --with-tdsver=7.0 configure
option. <<

Ok, I made the libraries with the tdsver option, and now the dsn-less
connection now works again.

Regarding the "timeout".  I looked at the 3 timeout tests and its not
clear to me whether they test for what I am experiencing.

What I am doing is in the middle of a query result set, I am physically
unplugging the network cable.  Like;

send sql statement
while not resultset eof
---here I pull the network cable after a couple of rows have been read
---get data
---check for an timeout error
---get the next row
loop

Interestingly, after appx 2 hours, an error is trapped, and Realbasic is
reporting the following error message it claims is from the odbc driver;

[FreeTDS][<drivername>]Communication link failure.

What do you recommend I do for my next step? 

Thanks!

Dan


On Wed, 2007-04-04 at 21:32 -0400, James K. Lowden wrote:
> Dan Mohn wrote:
> > I will note that with the pre-packaged
> > freetds, I was able to do a DSN-less connection, but with the same
> > setting I no longer seem to.  But that could just be my lack of
> > understanding.
> 
> Hello Dan,
> 
> When you connect DSN-less, you're relying on the built-in TDS version.
> Your package might have used the --with-tdsver=7.0 configure option. 
> 
> Glad to know things are working, though. 
> 
> > When
> > I disconnect from the network, the application still does not
> timeout.
> > However, I have noticed that if I wait 1 min, or 15 min, or longer
> and
> > re-establish the network connection, the application will then catch
> an
> > odbc error. 
> 
> There are timeout unit tests that bypass all the higher layers you're
> using.  Could I suggest you try the ODBC one?  If it passes but
> doesn't
> detect the kind of problem you're, well, creating, then we need a
> better
> test. 
> 
> Also please try my new bsqlodbc utility.  Then tell us precisely how
> you're testing: you issue query X, remove something, wait N seconds,
> replace something, then something happens.  Please try the same thing
> with
> bsqldb, because that will help tell us where the problem is, if any. 
> 
> I put a lot of time into this part of the project.  It's tricky, but I
> think I got it right.  If it's not right and you can show me, I'm at
> your
> service. 
> 
> Regards,
> 
> --jkl
> 
>