Message-ID: <f32941cee8bbe672bfa8ca3840795b6f8e11f24d.camel@cybertec.at>
Subject: Re: canceling statement coming in slave instance
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: Rohit Arora <arora.leo9@gmail.com>, pgsql-admin@postgresql.org
Date: Sat, 24 Nov 2018 20:05:18 +0100
In-Reply-To: 
 <CAMUAjH9FaYYJFkY6K5Exnb4ZA8KiPxOCqFCL9-AMTi+24ZYxdA@mail.gmail.com>
References: 
	<CAMUAjH9FaYYJFkY6K5Exnb4ZA8KiPxOCqFCL9-AMTi+24ZYxdA@mail.gmail.com>
Organization: Cybertec
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Precedence: bulk

Rohit Arora wrote:
> I occasionally encounter below error.
> 
> ERROR:  canceling statement due to conflict with recovery
> DETAIL:  User was holding a relation lock for too long
> 
> While investigating online i came to know that this issue can be handled by below configuration parameters
> 
> "max_standby_archive_delay "
> "max_standby_streaming_delay" 
> 
> I have increased the value of both the parameters as per below.
> 
> Current values:
> "max_standby_archive_delay=30s" ----> was increased it to 300s but we did not get any benefit so we rollback it to original value
> "max_standby_streaming_delay=300s"
> 
> But still i occasionally encounter the mention issue.

You'd have to set max_standby_streaming_delay to -1 to avoid the
error completely.  But bear in mind that it can delay replication
arbitrarily long.

Yours,
Laurenz Albe
-- 
Cybertec | https://www.cybertec-postgresql.com