Re: [patch] plproxy v2

From: Hannu Krosing <hannu(at)krosing(dot)net>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Marko Kreen <markokr(at)gmail(dot)com>, Andrew Sullivan <ajs(at)commandprompt(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: [patch] plproxy v2
Date: 2008-07-24 16:01:00
Message-ID: 1216915260.7001.53.camel@huvostro
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, 2008-07-22 at 11:25 -0400, Tom Lane wrote:
> "Marko Kreen" <markokr(at)gmail(dot)com> writes:
> > And user can execute only pre-determines queries/functions on system2.
>
> If that were actually the case then the security issue wouldn't loom
> quite so large, but the dynamic_query example in the plproxy regression
> tests provides a perfect example of how to ruin your security.

The idea is to allow the pl/proxy user only access to the needed
functions and nothing else on the remote db side.

dynamic_query ruins your security, if your pl/proxy remote user has too
much privileges.

> > Do you still see a big hole?
>
> Truck-sized, at least.
>
> The complaint here is not that it's impossible to use plproxy securely;
> the complaint is that it's so very easy to use it insecurely.

You mean "easy" like it is very easy to always use your OS as root ?

On Unix this is fixed by stating it as a bad idea in docs (and numerous
books), on windows you have a "privileged" checkbox when creating new
users.

---------------
Hannu

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Joshua D. Drake 2008-07-24 16:06:16 Re: Do we really want to migrate plproxy and citext into PG core distribution?
Previous Message Teodor Sigaev 2008-07-24 15:53:56 Re: [PATCHES] GIN improvements