Russell Smith <mr-russ(at)pws(dot)com(dot)au> writes:
> But this also means that a super user is not really a super user
> anymore. Are we comfortable with this redefinition?
Hm? It has never been the case that a view would act differently for a
superuser than it would when invoked by someone else. The bug here is
in ALTER OWNER failing to reassign the view's permissions checks
properly, not in how they are enforced later.
regards, tom lane
In response to
pgsql-bugs by date
|Next:||From: John R Pierce||Date: 2007-05-10 15:39:37|
|Subject: INSTALL appnote for Solaris 10...|
|Previous:||From: Russell Smith||Date: 2007-05-10 08:41:08|
|Subject: Re: BUG #3265: 8.1 -> 8.2 behviour change: View owner must
have access to underlying tables|