On Sun, Feb 13, 2011 at 10:19 PM, Selva manickaraja
<mavles78@gmail.com> wrote:
Dear All,
We had managed to setup and test the Continous Recovery of a secondary machine. What we need to achieve is for it to recover as primary in the event that primary fails. We know that we can use the "trigger_file" setting to fail over. However we understand that sometimes network connection can be intermittent and other factors which doesn't actually create a real "primary down" situation. So for those kind of scenarios can we set something in the configuration files like a time interval to wait before to fail-over. Has PostgreSQL this feature?
Postgres has no auto-failover. You would have to build that functionality yourself. Just make sure that (if you're actually going to automate it, which I don't usually recommend) your scripts are very pessimistic about failing over.
The trigger file is something that you would have to create either manually or (again, not recommended) programatically.
--Scott
Thank you.
Warmest Regards,
Selvan