| From: | "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com> |
|---|---|
| To: | Joe Conway <mail(at)joeconway(dot)com> |
| Cc: | Don Seiler <don(at)seiler(dot)us>, pgsql-admin <pgsql-admin(at)postgresql(dot)org> |
| Subject: | Re: Why is my table not autovacuuming? |
| Date: | 2017-08-28 17:33:01 |
| Message-ID: | CAKFQuwarsQaE80GjO0rz28Rhbt=vWo=-2s7T0_sMppAicm2_qg@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On Mon, Aug 28, 2017 at 9:44 AM, Joe Conway <mail(at)joeconway(dot)com> wrote:
> Then at
>
> some point your server lost power. The info in the stats collector gets
> reset on recovery.
Haven't pondered the vacuum side of the equation but ISTM if the
auto-analyze daemon is confronted with total amnesia that it should go
ahead and schedule an analyze on the entire database. ANALYZE is
relatively inexpensive.
And maybe learn some techniques to compensate. Like copying the most
recent snapshot of its input data to some place that survives a
crash-restart.
Also:
(formula) "is compared to the total number of tuples inserted, updated, or
deleted since the last ANALYZE."
https://www.postgresql.org/docs/devel/static/routine-vacuuming.html
Since there are other ways for the stats to reset beside running ANALYZE
this sentence hides a critical and should be amended to be more accurate.
David J.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Don Seiler | 2017-08-29 21:08:15 | Standby: Use WAL instead of Streaming for now |
| Previous Message | Joe Conway | 2017-08-28 17:31:05 | Re: Why is my table not autovacuuming? |