Biyi Akinpelu wrote:
The purpose of this exercise is to setup two database servers on one web server, one database runs being fully active, but if an error occurs along the way, the second server takes over. When the problem has been fixed on the main database server, its reverts back. Is it possible achieve this?
Hints: I am trying to a setup a database environment whereby I run both postgresql (main) and mysql (backup) database concurrently from an apache web-server. One acts as the main db server the other a backup of the main db server. In case any faults with the main db server, the backup automatically takes control of the server and reverts back when faults have been fixed.
So far I have managed to setup both databases. However, problem comes with automating the task to revert when error occurs and back again.
Any advice or suggestions will be deeply appreciated.
Thanx in advance.
How do I say "you're off your head," without seeming impolite? An application written to use postgresql won't use mysql without modification, You could write an application that will talk to either/both, but it would need to be fully tested against both. Every time it's changed. The two packages use different file formats, files created by one are inaccessible to the other. Last I looked, mysql lacked certain features essential to a relational database management system. If you use any of those standard features, you cannot use mysql. When I was making my decision as to which to write for, mysql stored money amounts as floating point, and worse, the mysql developers didn't understand what that is a bad idea. If you're interested in some technical discussion on the matter, the subject of mysql cmes up on the sql-ledger list sometimes. The author of mysql won't support mysql until it has the necessary features. Google a mysql-ledger list archive for these words: summerfield, browne, mysql, relational.