On Thu, Mar 29, 2007, Josh <[EMAIL PROTECTED]> said: > >> I assume that you're only running MySQL on one machine at a time, and >> using drbd to keep a hot copy of the block device? Or are you running the >> commercial MySQL cluster product? >> > The current setup is to use drbd to keep a hot copy of the block device. > I don't have enough RAM to hold the database in memory (the MySQL > cluster requires it all to be in RAM - no good for large db's)
Ok, so when the other side disappears, you turn off the drbd, then start MySQL and recover no differently than any other typical crash. Neat idea! Might I suggest just using MySQL's master-slave replication? It does the same thing but without the inherent risks of bringing up an imperfect block replica. [snip] > The reason to try drbd is that it looked fairly easy to fail over and > back, but now that I think of it a split-brain situation is going to > cause data loss no matter which way it's resolved. I'll have to do some > research on failing over MySQL and Postgres master-slave configurations. STOMITH :-P I just realized that DBMail can fix the IMAP UID problem after a split brain situation by reassigning id's and incrementing UIDVALIDITY. This would be way out in the future, though. So far we're trying very hard not to have to be hands-on about the underlying database replication. It will take lots and lots of code and tons of failure scenario test cases, so we won't start down that path until we're ready to commit to it. This is one of those things where half-assed code is far, far worse than no code at all. Aaron _______________________________________________ DBmail mailing list [email protected] https://mailman.fastxs.nl/mailman/listinfo/dbmail
