Обсуждение: Security Best Practices: Is This Reasonable?

I would appreciate any critique of this security model I want to use
for my planned web-accessible family database:

I have a working PostgreSQL running on a remote Linux web server
running Apache 2.  I want to allow web access to a database but want
to restrict row update to a row owner.

All access to the site is by SSL/TLS, and I use Apache htdigest
passwords to control access to the directory containing the database
Perl cgi scripts.

I plan to have every table have a field (called 'owner') which will be
filled in by the accessing user's name when a new row is created (I
will really use an integer key unique for each user).

The site user names and passwords will be the same as the database
user names and passwords.

I plan to have user names identified through the CGI environment and
then, for any attempted update of any row in any table, disallow it if
the user and owner do not match.

Thanks for any suggestions.

Best regards,

-Tom

We're doing something similar to allow dealers to access relevant parts of a table and not others. Using the Apache
username login works swell and if there's a security problem with it, we haven't bumped into it yet. 

Howard

Sent from my iPad (please disregard egg freckles)

On Jan 13, 2012, at 11:24 AM, Tom Browder <tom.browder@gmail.com> wrote:

> I would appreciate any critique of this security model I want to use
> for my planned web-accessible family database:
>
> I have a working PostgreSQL running on a remote Linux web server
> running Apache 2.  I want to allow web access to a database but want
> to restrict row update to a row owner.
>
> All access to the site is by SSL/TLS, and I use Apache htdigest
> passwords to control access to the directory containing the database
> Perl cgi scripts.
>
> I plan to have every table have a field (called 'owner') which will be
> filled in by the accessing user's name when a new row is created (I
> will really use an integer key unique for each user).
>
> The site user names and passwords will be the same as the database
> user names and passwords.
>
> I plan to have user names identified through the CGI environment and
> then, for any attempted update of any row in any table, disallow it if
> the user and owner do not match.
>
> Thanks for any suggestions.
>
> Best regards,
>
> -Tom
>
> --
> Sent via pgsql-novice mailing list (pgsql-novice@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-novice

On Fri, Jan 13, 2012 at 11:17, Howard Eglowstein
<howard@yankeescientific.com> wrote:
> We're doing something similar to allow dealers to access relevant parts of a table and not others. Using the Apache
username login works swell and if there's a security problem with it, we haven't bumped into it yet. 

Thanks, Howard.

Best regards,

-Tom