Check your documentation for backup and recovery.
Create what is called a stand by database. You copy and apply the archived
logfiles from the Prod to the stand by box and activate when a disaster
occurs.
Shaul <sbenyo_at_ibm.net> wrote in message news:7v5bju$gb0$1_at_news.ibm.net.il...
> Hi,
>
> Does anyone knows what's the best way to fully replicate a database ?
>
> I have a production database on one unix machine and I want to have a
> complete replica of this database one a second machine both on the same
LAN
> (100mbit).
>
> The replica database should be exactly the same as the production, so it
> will be able to replace it in case the other machine is burnt down.
>
> The replica database should be up-to-date with the production at all
times,
> minimally hurting the production's performace.
>
> If it's not possible to have an up-to-date replica without hurting
> performance, it is possible that the replica will be updated up to one
day
> back.
>
> 2 more important requirements:
>
> 1) The replica should be able to replace the production at any time with
> minimal work to do.
> 2) The replica should be activated at the shortest possible time.
>
> What's the best way of implementing such a replica ??
>
> I will be thankfull for any tips / help in this matter!
>
> 10x,
>
> Shaul
>
>
>
>
>
>
>
Received on Fri Oct 29 1999 - 08:07:50 CDT
