Re: Additional role attributes && superuser review - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: Additional role attributes && superuser review
Date
Msg-id 20160117185722.GF3685@tamriel.snowman.net
Whole thread Raw
In response to Re: Additional role attributes && superuser review  (Bruce Momjian <bruce@momjian.us>)
Responses Re: Additional role attributes && superuser review  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
* Bruce Momjian (bruce@momjian.us) wrote:
> On Sun, Jan 17, 2016 at 01:49:19PM -0500, Stephen Frost wrote:
> > * Bruce Momjian (bruce@momjian.us) wrote:
> > > > pgbackrest:
> > > >
> > > >   To run pgbackrest as a non-superuser and not the 'postgres' system
> > > >   user, grant the pg_backup role to the backrest user and ensure the
> > > >   backrest system user has read access to the database files (eg: by
> > > >   having the system user be a member of the 'postgres' group):
> > >                                 ------
> > >
> > > Just to clarify, the 'postgres' OS user group cannot read the data
> > > directory, e.g.
> > >
> > >     drwx------ 19 postgres staff 4096 Jan 17 12:19 data/
> > >         ^^^group
> > >
> > > I assume we don't want to change that.
> >
> > This is going to be distribution dependent, unfortunately.  On
> > Debian-based distributions, the group is 'postgres' and it'd be
> > perfectly reasonable to allow that group to read the data directory.
>
> Well, while the group name would be OS-dependent, the lack of any group
> permisions in not OS-dependent and is forced by initdb:
>
>     umask(S_IRWXG | S_IRWXO);
>
>     create_data_directory();

Right, we also check in the backend on startup for certain permissions.
I don't recall offhand if that's forced to 700 or if we allow 750.

> > I don't recall offhand if that means we'd have to make changes to allow
> > that, but, for my 2c, I don't see why we wouldn't allow it to be an
> > option.
>
> OK, that would be an initdb change then.

It would need to be optional, so distributions and users could choose
which makes sense for their systems.

Thanks!

Stephen

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Additional role attributes && superuser review
Next
From: "Constantin S. Pan"
Date:
Subject: Re: Proposal: speeding up GIN build with parallel workers