I'm finding that pg_dumps are dumping out, right near the end, the
following sequence of grants that are causing our QA folk a little bit
of concern:
REVOKE ALL ON SCHEMA public FROM PUBLIC;
REVOKE ALL ON SCHEMA public FROM chris;
GRANT ALL ON SCHEMA public TO chris;
GRANT ALL ON SCHEMA public TO PUBLIC;
The problem isn't anything terribly deep - it's just that there is no
user "chris" in their environment (chris happens to be one of the
superuser accounts on my workstation ;-)), so that the REVOKE/GRANT
combination raises errors.
Note that:
- I used the "postgres" superuser for anything needing superuserness
- I decline to use "sed" to filter this out; see the .sig ;-)
Is this an artifact of the fact that "chris" is the 'base superuser'?
--
output = ("cbbrowne" "@" "linuxfinances.info")
http://cbbrowne.com/info/postgresql.html
"Some people, when confronted with a Unix problem, think ‘I know, I’ll
use sed.’ Now they have two problems." - jwz@lucid.com