From: | Stephen Frost <sfrost(at)snowman(dot)net> |
---|---|
To: | "Campbell, Lance" <lance(at)illinois(dot)edu> |
Cc: | "pgsql-admin(at)postgresql(dot)org" <pgsql-admin(at)postgresql(dot)org> |
Subject: | Re: Partition tables in AWS RDS environment |
Date: | 2017-04-10 14:31:18 |
Message-ID: | 20170410143118.GK9812@tamriel.snowman.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Greetings Lance,
* Campbell, Lance (lance(at)illinois(dot)edu) wrote:
> In 9.6 if you do a pg_dump on a schema and you have a few very large tables in the schema will this cause the database to lock the table for writing?
In PG, readers do not block writers.
> This is an issue I ran into with older version of PostgreSQL using pg_dump on entire schemas. To get around this issue I partitioned the really large tables so that the locking would be a much shorter time period while running pg_dump. But this may not be necessary with 9.6
I'm not sure what issue you ran into, but pg_dump would block someone
from DROP'ing or TRUNACTE'ing a table (or other operations requiring a
very strong lock), not from simply doing an INSERT, UPDATE, or DELETE
(which take a lower lock that doesn't conflict with the locks pg_dump
takes).
Information about the various locks in PG is available here:
https://www.postgresql.org/docs/current/static/explicit-locking.html
Thanks!
Stephen
From | Date | Subject | |
---|---|---|---|
Next Message | Lazaro Garcia | 2017-04-10 15:09:42 | Help bad results with pgbench |
Previous Message | Campbell, Lance | 2017-04-10 14:25:13 | Re: Partition tables in AWS RDS environment |