Re: The Oracle ODBC driver story
Date: Fri, 11 Jun 2010 13:56:44 -0700 (PDT)
Message-ID: <1b142371-3b37-41b2-9fb4-582f3317b622_at_v12g2000prb.googlegroups.com>
On Jun 11, 12:43 pm, Mladen Gogala <n..._at_email.here.invalid> wrote:
> On Fri, 11 Jun 2010 09:56:56 -0700, joel garry wrote:
> > Why don't you ask Wim Coekaerts?
>
> I don't know the gentleman. I, of course, know of him but I don't want to
> pester people that I've never met with personal emails.
>
Fair enough. I imagine you aren't many degrees of freedom away from him, though - I see him publicly on linkedin, for example (I despise linkedin, but when I log in I see someone could introduce me to him, should I care, and he does say he shares expertise - I would go out on a limb and guess he wants to evangelize and make it work better). I imagine doing things through support adds degrees.
>
>
> > I suppose you could have rlib talk to something else with native links
> > that then talks to Oracle... or have a 10g intermediary... I'm assuming
> > you are on some 11g?
>
> Actually, no, I am not. The database in question is 10.2.0.5. The same
> problem happens with every version after 10.2.0.3. Version 10.2.0.3
> doesn't work because of some unresolved symbol on Red Hat 5.
And here I've been complaining about stuff that doesn't work on hp-ux because it has linux syntax, lol!
Oracle cross platform compatibility - makes all platforms red-headed stepchildren.
>
>
>
> > You mentioned compiling it from source? Isn't that the DIY way?
>
> Yes, it works that way, but I have to compile and configure the machine
> manually and my boss is reluctant to allow me to do that on the
> production box.
Fair enough. Any manual operations are subject to non-repeatability. I often even script things I only intend to do once.
jg
-- _at_home.com is bogus. Data and death: http://www.msnbc.msn.com/id/37612199/ns/us_news-life/Received on Fri Jun 11 2010 - 15:56:44 CDT