Skip to Content.
Sympa Menu

freetds - RE: [freetds] UDP, freeTDS and a v0.61 question

freetds AT lists.ibiblio.org

Subject: FreeTDS Development Group

List archive

Chronological Thread  
  • From: "Coleman, Dave" <DColeman AT US.TIAuto.com>
  • To: "'freetds AT lists.ibiblio.org'" <freetds AT lists.ibiblio.org>
  • Subject: RE: [freetds] UDP, freeTDS and a v0.61 question
  • Date: Fri, 31 Jan 2003 06:42:24 -0500

Thank you very much... we use MS SQL 7.0 on win 2K.

Thanks for clearing up the UPD thing.

> -----Original Message-----
> From: Frediano Ziglio [SMTP:freddyz77 AT tin.it]
> Sent: Thursday, January 30, 2003 3:03 PM
> To: FreeTDS
> Subject: Re: [freetds] UDP, freeTDS and a v0.61 question
>
> Il gio, 2003-01-30 alle 20:32, Coleman, Dave ha scritto:
> > Question for the group:
> >
> > Does the TDS specification include any mechanism for providing packet
> > redundancy? Obviously, the UDP protocol will not provide this.
> >
> > We are experiencing some strange errors on our system, however; these
> errors
> > are intermittent in nature.
> > * In the middle of some random batch (always the same script), we will
> > get the error, "Error converting NUMERIC to DECIMAL."
>
> Is hard to say without the code... perhaps NUMERIC and DECIMAL are
> stored in db in different way and have different range...
>
I'm not sure I should waste your time with that part of it. I was
simply hoping that someone would have had this error before and know how it
came about. It seems strange that something as consistant as SQL server
would glitch up like this.

> > * this same script (exact down to a strcmp()) will succeed
> > milliseconds later.
> > * The severity level of the error is 25, indicating that it is an
> > internal SQL server failure, not code, driver or query related.
> > * This raises an error on the client and everything that I can trace
> > to freeTDS seems to work just fine.
> > * The error code is 3601, which I cannot find anywhere.
> >
>
> I'll have a look. What server do you use? There is a syserrors table or
> similar...
>
Unfortunately, I've tried that. The server is MS SQL 7.0 and the
table is sysmessages. I think that at this point, I can safely point to the
network hardware or the SQL server configuration itself... Unless anyone
has a bright idea why the server would suddenly decide that it can't execute
the CONVERT function smack in the middle of run-time. I honestly don't
think this is a freeTDS issue, but I and my collegue are the only persons in
this organization with linux experience, so of course, it must be a problem
with the code until I prove otherwise. I'm sure more than a few people on
this list will understand how that comes to be.

> > With regards to the fact that TDS uses UDP:
> > * If I am utilizing only the tds layer of code, is there any
> > protection for corrupted/dropped network packets?
>
> TDS use TCP, not UDP, UDP is only used in discovery for mssql (still not
> implemented in FreeTDS)
>
This is very good news. Perhaps not a widely known fact, as every
IS person here was convinced that SQL used UDP and was very concerned when
they saw what was immediately labeled as "data corruption". Sorry for
having to mention it to clear the air.

> > * Can TCP be used from a DSN'less linux client to communicate with a
> > MS SQL server running on Win2k?
>
> If for DSN'less you mean ODBC DSN'less connection, yes, from some months
> ago (CVS version and coming 0.61). I use this configuration testing some
> win32 programs.
> If you mean use WINS instead of DNS for machine not in DNS I suggest
> configuring samba for WINS name resolution.
>
> ... omissis ...
>
Good. I ended up going the samba route. At least I know that was
the right way to do it.

> > Also, I am considering upgrading to 0.61 when released.
> >
> > My one concern is simple:
> > How much does the core tds API differ from 0.60 release 1? If I've
> > been following the list and keeping up to speed with changes, I'm going
> to
> > have more than a few structure alignment issues when I re-compile, or am
> I
> > wrong in this assumption? (Please tell me I'm wrong).
> >
>
> I think mainly differences are due to tds_process_XXX functions and
> adding of others... Also is absolutely not binary compatible...
>
I was sooooo afraid you would say that, but thanks for the warning,
it's better than finding out with a screen full of compile errors. :-)

> Regards,
> Frediano Ziglio
>
>
> _______________________________________________
> FreeTDS mailing list
> FreeTDS AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/freetds




Archive powered by MHonArc 2.6.24.

Top of Page