I was trying to design it so that the slaves wouldn't know they had connected to a different master, as they both masters would have the same IP address that gets failed over based on the Linux Virtual Server software and VRRP (like heartbeat from Linux-HA).
I'm beginning to think that my configuration is too complex, and would introduce too many places for failure. Off hand do you know any good resources on how to setup a highly available and load balanced MySQL system? Setting up replication is easy enough, but I can't find much data about getting the whole thing to work together so to speak (highly available writes). Thanks! ....................... Ian Neubert Director of IS TWAcomm.com, Inc. http://www.twacomm.com/ -----Original Message----- From: Dathan Vance Pattishall [mailto:[EMAIL PROTECTED] Sent: Thursday, August 07, 2003 12:54 PM To: 'Ian Neubert'; [EMAIL PROTECTED] Subject: RE: MySQL Replication That configuration doesn't seem to work if I'm reading it right. A slave cannot have more then one master unless mysql more then 1 mysqld process running on different ports is used. -->For example, if a person places an order on our site, the update is sent -->to -->the master server. Can that update be replicated to the slaves before the -->browser is taken to a reciept page (which is queried against the slaves)? Also guaranteeing an event has occurred on a slave and building application logic dependent on this necessity is a tough problem to solve and you might as well query the master since you need to verify the event is there anyway as part of the conditional. The reason for my last statement, I've notice in 3.23.5x that replication lag can occur in many situations: - The master is loaded and is not able to stream the binary log file as fast. - Locks on the slave preventing updates to occur - Load on the slave is high - Invalid packet errors where the slave needs to reconnect - Some sort of error that stops replication With all of the above examples, I would not have application logic dependent on slave events. Instead code your applications based on the known limitations. For example a search engine. It does not necessarily need an up to date all of the time index of words and matches... -- Dathan -->-----Original Message----- -->From: Ian Neubert [mailto:[EMAIL PROTECTED] -->Sent: Thursday, August 07, 2003 11:27 AM -->To: [EMAIL PROTECTED] -->Subject: MySQL Replication --> -->Hello all, --> -->Can anyone share their experience with replication? In particular I'm -->wondering how fast (or how slow?) replication occurs. Could I pull off -->the -->configuration below? --> --> -->http://linux.ianneubert.com/images/mysql_failover.png or -->http://linux.ianneubert.com/images/mysql_failover.pdf --> -->Thanks for your insight! --> -->....................... -->Ian Neubert -->Director of IS -->TWAcomm.com, Inc. -->http://www.twacomm.com/ --> --> -->-- -->MySQL General Mailing List -->For list archives: http://lists.mysql.com/mysql -->To unsubscribe: -->http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
