Re: changing sort_mem on the fly?

On Thu, 2005-01-27 at 20:03 -0600, Jim C. Nasby wrote:
> Yes, there is a risk allowing ad-hoc settings; you can starve the
> machine for memory.

A malicious user who can execute SELECT queries can already consume an
arbitrary amount of memory -- say, by disabling GEQO and self-joining
pg_class to itself 50 times. I'm not sure that letting users modify
sort_mem/work_mem actually increases the risk from malicious users.
Restricting this parameter to superusers only would also be a hit to
usability.

-Neil