On Saturday 02 May 2009 13:08, Daniel Verite wrote:
> Terry Lee Tucker writes
>
> > Q1: Can we set up a scenario where there is more that one
> > warm standby?
>
> Yes. But you'll have to consider what you want to happen when one
> standby is correctly receiving the WAL files and another is not,
> because the archive_command has to either fail or succeed, it can't
> return "half-done".
> You may need to provide your own layer that retry pushing local copies
> of WAL files to the remote nodes and deleting them only when they've
> been received by every standby server.
>
> > Q2: Am I correct in assuming that ALL changes to any of the
> > production schema will be written to the warm standby?
> > For example, if I drop a constraint in production I assume
> > the same will occur on the warm standby. If I create and
> > drop a table in production, I assume it will occur on the warm
> > standby.
>
> It will, all DDL is replicated.
>
> Best regards,
Daniel:
Thanks for the reply. I later read the answer to the first question in the
docs. I expected the answer to number 2 to be affirmative, but just wanted to
make sure.
Thanks for the help...
--
Terry Lee Tucker
Turbo's IT Manager
Turbo, division of OHL
2251 Jesse Jewell Pkwy
Gainesville, GA 30501
tel: (336) 372-6812 cell: (336) 404-6987
terry@turbocorp.com
www.turbocorp.com
