From: | Chris Withers <chris(at)withers(dot)org> |
---|---|
To: | pgsql-general <pgsql-general(at)postgresql(dot)org> |
Subject: | debugging intermittent slow updates under higher load |
Date: | 2018-12-05 14:34:18 |
Message-ID: | bda6cf39-08ad-322f-d562-49c042cb4012@withers.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Hi All,
This is on postgres 9.4.16, same table as the last question I asked,
here's an abbreviated desc:
# \d alerts_alert
Table "public.alerts_alert"
Column | Type | Modifiers
-----------------+--------------------------+-----------
tags | jsonb | not null
id | character varying(86) | not null
...
Indexes:
"alerts_alert_pkey" PRIMARY KEY, btree (id)
The table has around 1.5M rows which have been updated/inserted around
121M times, the distribution of updates to row in alerts_alert will be
quite uneven, from 1 insert up to 1 insert and 0.5M updates.
Under high load (200-300 inserts/updates per second) we see occasional
(~10 per hour) updates taking excessively long times (2-10s). These
updates are always of the form:
UPDATE "alerts_alert" SET ...bunch of fields... WHERE
"alerts_alert"."id" = '...sha1 hash...';
Here's a sample explain:
https://explain.depesz.com/s/Fjq8
What could be causing this? What could we do to debug? What config
changes could we make to alleviate this?
cheers,
Chris
From | Date | Subject | |
---|---|---|---|
Next Message | Stephen Frost | 2018-12-05 14:38:53 | Re: surprising query optimisation |
Previous Message | Stephen Frost | 2018-12-05 14:33:40 | Re: surprising query optimisation |