From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Justin Pasher" <justinp(at)newmediagateway(dot)com> |
Cc: | "'Richard Huxton'" <dev(at)archonet(dot)com>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: Fixing broken permissions for deleted user |
Date: | 2007-05-18 04:15:30 |
Message-ID: | 19335.1179461730@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Justin Pasher" <justinp(at)newmediagateway(dot)com> writes:
> Perfect. Just was I was looking for. So is it safe to actually run an update
> on the pg_catalog.pg_type.typowner column to change the user id from 101 to
> another existing user id without causing any other database weirdness?
Should work. In recent PG versions you would need to worry about
pg_shdepend entries too, but if you had pg_shdepend you would not have
been able to get into this state in the first place (in theory anyway).
Some general suggestions about manual changes to the system catalogs:
Reasonable prudence would suggest making the change inside a BEGIN block
and looking around for problems before you COMMIT. If you're really
paranoid, you could first stop the postmaster and take a plain tarball
backup of the PGDATA directory tree, which would certainly let you get
back to where you were if things go horribly wrong.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Albe Laurenz | 2007-05-18 07:26:09 | Re: HowTo SSL probaply with ODBC ? |
Previous Message | Justin Pasher | 2007-05-18 04:05:13 | Re: Fixing broken permissions for deleted user |