Re: BUG #3947: Unable to Initialize Data Cluster

From: "Dave Page" <dpage(at)pgadmin(dot)org>
To: "Fusion Software (UK) Ltd" <lee(at)fusionsoftwareuk(dot)co(dot)uk>, pgsql-bugs(at)postgresql(dot)org
Subject: Re: BUG #3947: Unable to Initialize Data Cluster
Date: 2008-02-10 17:50:30
Message-ID: 937d27e10802100950s69683150ue16ba5e53d815ee5@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Feb 10, 2008 5:10 PM, Fusion Software (UK) Ltd
<lee(at)fusionsoftwareuk(dot)co(dot)uk> wrote:
>
> Hi Dave
>
> Is this machine on a domain with any group policies or similar? I assume
> you're running the installations as Administrator?
> Its a basic default Vista installation with no domain or group policies.
> Most of our users just buy a PC and use it straight out of the box which is
> what we're trying to imitate. Yes, we're installing everything as a
> administrator.

OK. How did you install? I've seen some odd issues recently with
unattended installations of Vista.

> The most concerning thing is that it works on one PC and not another, even
> though they're both clean Vista installations, nothing else on them, albeit
> the working one is Vista Business and the other is Vista Home Premium. I
> have to say though that the pgSQL service really screws things up on that
> PC, it takes an absolute age to open the services screen once it's
> malfunctioned and the whole system freezes for several minutes until it
> finally opens the services showing that it hasn't started.

Hmm...

> Did you try any of the beta or RC releases?
> No, we deliberately waited until the final release so that we wouldn't run
> into exactly the issues we've just run into....:-(

D'oh! This is open source - we need people like you to be testing the
pre-release versions to help us nail any issues that crop up before we
release the GA version.

> Quite understood - obviously we're pretty keen to fix it as well.
> I know you are and don't get me wrong. Since our discovery of PostGresql
> about a year ago we have been nothing but impressed, and would certainly not
> look anywhere else for a superb database backend. I just hope in a way that
> this anomaly occurs all over the place and becomes reproducable. There's
> nothing worse than trying to fix a bug you can't replicate!

No, absolutely not :-(

A couple of quick thoughts - does the problem go away if you copy the
runtimes into the PostgreSQL bin directory? The merge module should
install them into
C:\windows\WinSxS\x86_Microsoft.VC80.CRT_1fc8b3b9a1e18e3b_8.0.50727.762_x-ww_6b128700

Also, do you even see the runtimes in that directory following the
installation? What about corresponding manifests in
C:\windows\WinSxS\Manifests?

--
Dave Page
EnterpriseDB UK: http://www.enterprisedb.com
The Oracle-compatible database company

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2008-02-10 18:34:48 Re: 8.3.0: vacuum full analyze: "invalid memory alloc request size"
Previous Message Tom Lane 2008-02-10 17:19:40 Re: BUG #3950: Server Problems