At 08:14 2002-03-14, Konstantin Tsolov wrote: >right, of course ;-) (i neglected the join part) > >so, now i know exactly whom to ask ;-) > >i have the following bother (it's in the planning state): > >i have the idea of setting up a two-way replication system with 2 mysql >servers for my radius database. >i need them to act interchangeably: > >1. if server1 is alive, all selects and updates (radius acct) should go to >server1. server2 should act as a slave for server1. > >2. if server1 dies, all selects and updates (radius acct) should go to >server2. > >so far, so good. >but here comes the tricky (for me) part - when server1 comes up again, it >shold do 2 things: >a) catch up with server2. >b) become master again, leaving server2 as it's slave (same situation as in >1. above). > >how could this happen ? >all ideas are appreciated (pointing to reading - also).
Try here, since You are already thinking of using 2 machines: http://linux-ha.org/LinuxFailSafe/ Works rather well with IP, MAC and Oracle/SAP failover. I believe there is a driver for MySQL, if not You shouldn't t have much trouble writing one, examples are provided. It's all open source, same code _base_ as IRIX FailSafe (HA on SGI machines for the past 7? years). Since this is really off-topic, if You have any further questions, You probably should ask me off the list. I would only suggest using either DRBD or shared-SCSI for storage. --------------------------------------------------------------------- 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
