Mohammed Rashid wrote:
Hi All, I think there is some confusion regarding this. I want postgres to be installed at two different PCs.Then the data from each postgres should be automatically updated to the third PC which we call server. This process should be automatic. It should also take care of network disconnection between client and server. After establishment of network connections the remaining data should be automatically updated to the server. It is bit different from the mirroring of databases, but the concept is same. I just want to ask whether the Postgres database supports this or not? Or I have to write an application to do so.
I think this sounds like you want a server in the office and the two clients are maybe laptops that want to synchronise changes between their own data and the office server when they come into the office. So they would send changes to the server as well as receive changes from the server. This case of intermittent network connections isn't usually handled by replication options so would be a custom setup by you. To automate the task you want some sort of heartbeat monitor to detect the presence of the server and perform the synchronising. You will either want a list of changed records per client or modified timestamps to determine what has changed since the last synchronise, as well as define actions upon conflicts. -- Shane Ambler pgSQL (at) Sheeky (dot) Biz Get Sheeky @ http://Sheeky.Biz
