Re: Data Redundancy

[Tom Brown <tom@nextstate.net>]

On Mon, 2007-01-22 at 15:43 +1000, Will Parkinson wrote:
> Hi All,
> 
> I have a server that i need to mirror elsewhere for data redundancy 
> purposes.  What is the best way to have this done?  The sites on my box 
> are MySQL and PostgreSQL driven so those need to be on both machines as 
> well.  Ideally id like to have a main server and in any event that the 
> main server went down, switch to the other server. 

That's exactly what drbd+heartbeat does for you. drbd mirrors your data
and heartbeat handles the fail over. (see drbd.org and linux-ha.org)
This has worked for me. I think I saw someone mention heartbeat and
linux-ha already, so sorry if this is redundant. The ideal scenario is
to have both machines identical, but that is not necessary. So, if you
have another server you can use, it may be cheap that way. However,
there is a time commitment to learning how to use the two tools.

Good luck,
Tom