Re: dblink vs SQL/MED - security and implementation details - Mailing list pgsql-hackers

From Tom Lane
Subject Re: dblink vs SQL/MED - security and implementation details
Date
Msg-id 9169.1231266670@sss.pgh.pa.us
Whole thread Raw
In response to Re: dblink vs SQL/MED - security and implementation details  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
Peter Eisentraut <peter_e@gmx.net> writes:
> On Tuesday 06 January 2009 19:50:51 Tom Lane wrote:
>> What about the permissions on the system catalogs themselves?
>> AFAICT, the pg_user_mappings view will expose user passwords to
>> the "owner" of the foreign server, which doesn't seem good.

> Well, no one is forcing you to put a password there.  dblink has had its 
> mechanisms for obtaining passwords until now, and those are not invalidated 
> by this.  There are as always limited use cases for hardcoding passwords, but 
> in a fully multiuser environment you probably want to use a different 
> authentication mechanism.  Eventually, when we allow these modules to 
> actually call out, we will have to seriously evaluate that.  But for right 
> now, if you don't want your password in there, don't put it there.

Huh?  The advertised reason for putting in all this stuff was to provide
a thought-through, secure mechanism for dealing with connection
information.  If we haven't done that thinking yet, I'm of the opinion
the whole thing should be ripped out until we have.  It's of exactly
zero value if it cannot be trusted with a password.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: dblink vs SQL/MED - security and implementation details
Next
From: Magnus Hagander
Date:
Subject: Re: version() output vs. 32/64 bits