Thread: Writing to a database or schema on a slave
Hey everyone. I'd like to create a schema on my slave so that users who do not have access to the master can create some data. Clearly this data won't be replicated, since it's on the slave, but will it cause any problems w/ data integrity to have it on the slave?
--
Wells Oliver
wellsoliver@gmail.com
Wells Oliver
wellsoliver@gmail.com
Wells Oliver <wellsoliver@gmail.com> wrote: > I'd like to create a schema on my slave so that users who do not > have access to the master can create some data. Clearly this data > won't be replicated, since it's on the slave, but will it cause > any problems w/ data integrity to have it on the slave? What are you using for replication? -Kevin
I am using the WAL method w/ 9.1.
--
Wells Oliver
wellsoliver@gmail.com
On Tue, Apr 10, 2012 at 4:29 PM, Kevin Grittner <Kevin.Grittner@wicourts.gov> wrote:
Wells Oliver <wellsoliver@gmail.com> wrote:What are you using for replication?
> I'd like to create a schema on my slave so that users who do not
> have access to the master can create some data. Clearly this data
> won't be replicated, since it's on the slave, but will it cause
> any problems w/ data integrity to have it on the slave?
-Kevin
Wells Oliver
wellsoliver@gmail.com
[rearranged; please don't top-post] Wells Oliver <wellsoliver@gmail.com> wrote: > Kevin Grittner <Kevin.Grittner@wicourts.gov wrote: >> Wells Oliver <wellsoliver@gmail.com> wrote: >> >>> I'd like to create a schema on my slave so that users who do not >>> have access to the master can create some data. Clearly this >>> data won't be replicated, since it's on the slave, but will it >>> cause any problems w/ data integrity to have it on the slave? >> >> What are you using for replication? > I am using the WAL method w/ 9.1. Then the replica must be read-only; you can't create anything on that PostgreSQL instance because it is a block-level copy of the master database cluster. You would need to use logical replication (for example, Slony) to do what you describe, or store the new schema on another cluster and use dblink or a foreign data wrapper. -Kevin