Re: OPS, Failsafe, Standby...

OPS - lose a server keep running, new connections established through remaining severs.
OFS - NT solution, Short outage while failing server shuts down and starts oracle on second node, new connections established after second node initialized. With properly coded application can roll connection over to second node, I believe. Standby - a copy of your database running on a second server which constantly is applying archive logs. To roll your appl over to this standby server, or at least what we were planning to do, we would change the IP address on the standby node to that of the primary node (which died).
Terry

Samuel Gay wrote:

> Hi,
>
> We will begin a new project, we need an application wich run 24 hours/24, and 7
> days/7. Then we need 2 servers, with separates disks, to avoid hardware failure,
> ... If the first server (or database on it) crash, we need to have on the second
> serveur data from the first (or at least data exepct last 10 seconds). The delay
> to activate the second server have to be short (1-2 minutes maxi).
>
> We will use Linux server, it's more reliable then NT, and have a moderate cost.
>
> If somebody use already a solution to solve a similar problem, I like to have
> informations.
>
> Thanks in advance.
>
> Samuel GAY.
>
> ------
> User of http://www.foorum.com/. The best tools for usenet searching.
  Received on Wed Sep 13 2000 - 22:27:47 CDT