Hi, I am about to set up a network which has a central "main" server, and several (up to 30) "shop" servers (it is for a retail chain). I'm intending to link install mysql on all of them and link them up using the method described below. What I'd like to know is (a.) Can anyone see a fatal flaw in my plan, (b.) Can anyone suggest a better way), and (c.)Is there anything i should watch out for ?
Requirement: Certain tables in the database must be replicated round all of the shops, so that changes e.g. in prices made on a central server can cascade out the shop servers, and also changes to stock levels, messages etc made in one shop are copied to the other shops and the main server. The main server will always be online, and the shops will connect to the Internet via DSL or Dialup/ISDN. Here's the plan: I will configure each of the servers to be both a master and a slave (same as for normal two way replication betweem to machines) In the slave settings on the shop servers, the "master" will always be set to the main central server. I will have a script running on the main server that executes a "CHANGE MASTER TO" each of the shop servers periodically. This way, in theory, the changes will propagate round whenever the cycle is completed. I'd appreciate your thoughts! Thanks in advance, Iain Gray --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php