| From: | Christopher Kings-Lynne <chriskl(at)familyhealth(dot)com(dot)au> |
|---|---|
| To: | Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Peter Eisentraut <peter_e(at)gmx(dot)net> |
| Subject: | Re: OWNER TO on all objects |
| Date: | 2004-06-16 05:41:35 |
| Message-ID: | 40CFDD8F.1020404@familyhealth.com.au |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> - How does the above point affect full dumps that include schema and
> data? In my proposal, the copy commands will run as the user running
> the script, not the table owner anymore. Presumably, the user running
> the script is a superuser. Given that it is possible for a table owner
> to revoke their own INSERT privilege on their table, the existing
> behaviour is broken anyway.
Yes, confirmed. Current pg_dump is quite broken in the situation where
a table owner revokes their own INSERT privilege. It attempts to run
COPY as the owner of the table. Only a superuser is guaranteed to be
able to always insert. We should fix that.
Chris
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Christopher Kings-Lynne | 2004-06-16 06:43:36 | Re: OWNER TO on all objects |
| Previous Message | Christopher Kings-Lynne | 2004-06-16 04:53:41 | Re: OWNER TO on all objects |