On 02/24/2011 10:35 AM, Andrea Brancatelli wrote: > Pardon me but if you write only on one node the it's not multi master > :-)
Even if it's not *used* as multi-master, it can still be *configured* as a multi-master. > > Btw, never tried with dbmail but, in general, I have at least two > cluster instances of multi master multi writers running since almost > 2 years without any particular issues. In one case the replicated > cluster is distributed geographically, with two node in a datacenter > for wider use and one node in the office for local use. It mainly > serves web applications written in php and the load is quite high > (and the db is pretty big as well, almost 80gb), although it's > clearly different from dbmail's db usage pattern. So? Replication in mysql is very robust. Never meant to imply anything else. > There are things you have to watch out, like auto increment offset > and such, but it's not rocket science... (and have an automated > monitoring script that halts everything if replication goes out > sync!!) Auto-increment offsets will prevent key-collisions, but offsets will also break imap compliance because the UIDNEXT value may end up being lower than the next inserted UID, leading to missing messages in the MUA. There is *no* known workable solution to this problem short of major rework of the imap codebase. -- ________________________________________________________________ Paul Stevens paul at nfg.nl NET FACILITIES GROUP GPG/PGP: 1024D/11F8CD31 The Netherlands________________________________http://www.nfg.nl _______________________________________________ DBmail mailing list [email protected] http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
