From: | Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com> |
---|---|
To: | stan <stanb(at)panix(dot)com>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: A question about sequences and backup/restore cycles |
Date: | 2019-10-22 00:24:37 |
Message-ID: | c2a2c79f-b004-81b7-9a51-1fbc8bc341e2@aklaver.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On 10/21/19 5:13 PM, stan wrote:
>
> I typically design a system with primary keys defined, like this:
>
>
> CREATE TABLE employee (
> employee_key integer DEFAULT nextval('employee_key_serial')
> PRIMARY KEY ,
>
> I use scripts to build the database structures and load the data. I am
> careful to get the dependencies in the correct order, so that the keys later
> structures depend on already exist.
>
> Today I was going over the design for the current project with a friend,
> whose expertise i respect. he said that he had issues on a system designed
> by an OEM that had dependencies on keys developed from sequences after a
> backup/restore cycle,
>
> Will I potentially have these issues? If so, what can I do different to
> avoid this being an issue?
It is not clear to me what you are doing:
1) Are you using pg_dump/pg_restore to populate a database?
If so it will take care of the dependencies.
2) Are you using a home built method to populate the database?
In that case you take responsibility for dependencies.
>
>
--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com
From | Date | Subject | |
---|---|---|---|
Next Message | Maciek Sakrejda | 2019-10-22 06:18:32 | EXPLAIN BUFFERS and I/O timing accounting questions |
Previous Message | stan | 2019-10-22 00:13:31 | A question about sequences and backup/restore cycles |