quick question about mysql replication

nephish

hey there,
i have a debian linux web server and another backup computer that is set up almost identical to it.
i was looking at the mysql function of replication, to have the main computer as the master, and the other as the slave.
my question is, is this for mirroring only ? or could i have them both take data in independantly of each other and only syncronize about once or twice / day ?

or is this even the most reccomended way to do a data sync of two identical databases ?
i basically need one that is as up to date with the other in case the main one dies., so i can just change its internal ip and make it the web server.

open to all suggestions,
thanks

Roger_Ramjet

Short answer is NO. I don't know if ver 5 allows for replication sets, but ver 4 is only master/slave replication, ie mirroring, and updates can only be replicated from the master to the slaves.

nephish

ok ,yeah, i am using ver 4 so, thats out. what would be a good way to collect data on both servers all day, then like at midnight or so, syncronize the two databases, giving preference to the main server on any discrephencies ? anything simple ?
thanks

Roger_Ramjet

I've had a good read of the manual and found that it does mention two-way replication as a possibility, but give no information as to how.

What I did find is mysql clustering which may be more of what you are after. In fact, I'm certain it is the solution you really need.

nephish

thanks, i have been looking over it. i may try to implement it when i get a chance to do it very step by step, looks a little complicated. but not impossible

thanks a lot.