From: | Stephen Frost <sfrost(at)snowman(dot)net> |
---|---|
To: | John R Pierce <pierce(at)hogranch(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Tablespace limit feature |
Date: | 2014-11-04 17:46:11 |
Message-ID: | 20141104174611.GC28859@tamriel.snowman.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
* John R Pierce (pierce(at)hogranch(dot)com) wrote:
> On 11/4/2014 9:00 AM, Alejandro Carrillo wrote:
> >I need to limit the tablespace file because I need to control the
> >hard disk space used in a tablespace
>
> so what happens if a query does something that violates the quota?
It gets cancelled. See temp_file_limit.
> I can see one difficulty in implementation would be just keeping
> track of how much space is used in a tablespace, thats not done
> right now at all as far as I know. every time a new block is
> allocated, a tablespace size catalog table would have to be
> updated.... and oh no, what happens if THAT table grows and
> overflows the quota of the default tablespace where the catalog is
> maintained?
There are a bunch of questions, but I doubt we'd apply the quota to
system catalogs regardless. That does lead to an interesting question
about quotas for number of tables, or indexes, or whatever, but I tend
to doubt that's as much of an issue. Such a quota isn't to address
individuals who are actively attempting to attack the system (there's
lots of ways to cause problems if you've got SQL-level access) but
rather just to manage space limitations among different individuals
using the system.
Thanks,
Stephen
From | Date | Subject | |
---|---|---|---|
Next Message | Bernd Helmle | 2014-11-04 18:23:21 | Re: COPY TO returning empty result with parallel ALTER TABLE |
Previous Message | John R Pierce | 2014-11-04 17:36:00 | Re: Tablespace limit feature |