The current Jabber (and shared with the IM world) is based upon the normal Internet model of 1 client, 1 connection to 1 server. But that server is really just an IP address. So when you ask yourself what do I do when my Jabber server goes down, it's the same question as "what do I do if I lose my mail or Web server". The answer is to move your single point of failure out from your Jabber server. Setting up base Jabber to handle load-balancing and redunandt servers wouldn't be too much of a problem. Doing things like groupchat might because I'm not sure if there is a facility to share conference information between servers. For example u1,u2,u3 are on J1 and u4,u5,u6 are on J2 but all should be in Conference C1. This can fairly easily remedied by moving groupchat from a shared module (ie loaded onto each server instance) to a server module 1 instance shared by the jabber servers. Of course you still must have a redunandt system for the conference server. So in theory it's quite possible to build redunandt Jabber servers. However, the practicality of it, is going to be difficult for now. Mark Mark Wilcox [EMAIL PROTECTED] Got LDAP? _______________________________________________ jdev mailing list [EMAIL PROTECTED] http://mailman.jabber.org/listinfo/jdev
