On Fri, 25 Jul 2003 11:09 pm, Helen Borrie wrote:

> At 07:56 AM 26/07/2003 +0000, you wrote:
> >heLen,
> >
> >I meant is there a FreeBSD binary of 1.0.0.796-0 specifically?. I have
> > found the 1.0.2 binary,
>
> I'm pretty certain that 1.0.2 IS build 796, i.e. the tagged build that
> IBPhoenix produced at Christmas for Windows and Linux and other platform
> maintainers built at around the same time. The only released build for
> 1.0.x since then was 1.0.3., and it was much more recent (beginning of
> June). There hasn't been a 1.0.3 build for FreeBSD at all.
>

Excellent, I will give 1.0.2 a go using the native binaries and see how we go.

> >but as mentioned in my last post, there's a huge amount of
> >politics if I run any other version bar the mentioned one. It all seems
> >really stupid to me. I think I might just give the 1.0.2 a whirl anyway
> > and try convince the software company to support it.
>
> Just unpack the compressed file and look at the version number. As I
> recall, there was a small stuff-up, so the server came out with a build
> number of 796 while the client was 798. I don't think you have anything
> to worry about, but I wonder if your current problems have anything to do
> with having the wrong client library. Have you checked that?
>

Well, being that the "real" client for this thing is actually the software
written by this company (which is how I noticed the problem, I only verified
it using the isql client), no I haven't sepcifically checked that. I will opt
for the native binary of 1.0.2 and let you know how I go.

> Incidentally, AFAIK, nothing fundamentally changed in the 1.0.x builds
> since the original 1.0 release that affects the way clients connect. The
> big change happened during the release candidate phase, more than a year
> ago.
>

If nothing has fundamentally changed (which I assumed it hadn't, but wasn't
100% on that), then that helps my case against this company to in fact use a
native 1.0.2 release as well. Thank you, that information alone is very
useful.

> To update you on my tests with RC 1.5, I continue to observe the same
> phenomenon as you do with Classic server on Red Hat 8. It appears to me
> that the only user who can log in more than once using direct local
> connection is root, because root can do whatever it likes. SYSDBA can
> make exactly one local connection and thereafter any number of localhost
> connections; and I can log any user in from any other station on the
> network.
>
> Don't expect Classic server on *x to behave exactly like SuperServer on
> Windows. There is a vast architectural difference. I really don't think
> your problem is a problem, I think it is "as designed".
> heLen
>

Oh I don't expect CS to behave the same as superserver at all, I used CS on
Win 32 (1.0.0.796-0) and CS on *nix when comparing. My understanding, though
new to firebird is that CS is very different to SS, and I know I cannot go
outside the realm of CS for this project, the software written relies on it
beingCS (I have been told).

Thankyou very much for your help thus far, I really appreciate it.