| From: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com> |
|---|---|
| To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
| Cc: | Bruce Momjian <bruce(at)momjian(dot)us>, Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Grigory Smolkin <g(dot)smolkin(at)postgrespro(dot)ru>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Fun fact about autovacuum and orphan temp tables |
| Date: | 2016-10-21 05:16:36 |
| Message-ID: | CAB7nPqRb7jvj2kHT7oeEOaL7R_+6fi3ZMNrV1=Ru7p0gBbKogw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Sep 8, 2016 at 12:38 AM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> On Mon, Sep 5, 2016 at 1:14 PM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
>> I don't think we look at those temp tables frequently enough to justify
>> keeping them around for all users.
>
> +1. I think it would be much better to nuke them more aggressively.
+1 from here as well. Making the deletion of orphaned temp tables even
in the non-wraparound autovacuum case mandatory looks to be the better
move to me. I can see that it could be important to be able to look at
some of temp tables' data after a crash, but the argument looks weak
compared to the potential bloat of catalog tables because of those
dangling temp relations. And I'd suspect that there are far more users
who would like to see this removal more aggressive than users caring
about having a look at those orphaned tables after a crash.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Michael Paquier | 2016-10-21 05:29:24 | Re: Fun fact about autovacuum and orphan temp tables |
| Previous Message | Craig Ringer | 2016-10-21 04:50:43 | Re: Remove autovacuum GUC? |