From: | Israel Brewster <israel(at)ravnalaska(dot)net> |
---|---|
To: | pgsql-general <pgsql-general(at)postgresql(dot)org> |
Subject: | pg_dump Conflict with recovery |
Date: | 2017-01-23 18:24:06 |
Message-ID: | 6A99621F-F35E-4335-97EA-C4449B030179@ravnalaska.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
I have a backup strategy that in part consists of doing pg_dump s on my various databases. In order to hopefully reduce/prevent operational slow-down as a result of the backup, I do the dumps from my secondary server, configured as a hot standby with streaming replication.
In general this works fine, but one of my databases has now grown to the point that often as not I get the following when trying to dump the database:
ERROR: canceling statement due to conflict with recovery
DETAIL: User was holding a relation lock for too long.
As I understand it, this is due to the pg_dump taking longer than the max_standby_streaming_delay of 180s, and as such could be easily fixed by upping that value in the config. But is that the "right" fix? Or is there a "better" way?
-----------------------------------------------
Israel Brewster
Systems Analyst II
Ravn Alaska
5245 Airport Industrial Rd
Fairbanks, AK 99709
(907) 450-7293
-----------------------------------------------
From | Date | Subject | |
---|---|---|---|
Next Message | Jerry Sievers | 2017-01-23 19:00:33 | Re: Why does this hot standy archive_command work |
Previous Message | btober@computer.org | 2017-01-23 18:06:01 | Re: Why does this hot standy archive_command work |