Re: Unhappy thoughts about pg_dump and objects inherited from template1 - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Unhappy thoughts about pg_dump and objects inherited from template1
Date
Msg-id 6439.973785029@sss.pgh.pa.us
Whole thread Raw
In response to Re: Unhappy thoughts about pg_dump and objects inherited from template1  (Jan Wieck <janwieck@Yahoo.com>)
List pgsql-hackers
Jan Wieck <janwieck@Yahoo.com> writes:
>     For now we have oid's 1-16383 hardwired from the  bki  files.
>     Some  16384-xxxxx get allocated by initdb after bootstrap, so
>     we just need to bump the oid counter at the end of initdb (by
>     some  bootstrap  interface  command)  to  lets  say 32768 and
>     reject any attempt to touch an object with a lower oid.

What do you mean by "touch"?  The system catalogs certainly can't
be made read-only in general.

AFAIK we already have sufficient defenses against unwanted hackery on
the system catalogs, and so I don't really see a need for another level
of checking.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Question about reliability?
Next
From: Zeugswetter Andreas SB
Date:
Subject: AW: AW: Unhappy thoughts about pg_dump and objects inhe rited from template1