https://www.ejabberd.im/node/3794 en https://www.ejabberd.im/node/3794#comment-55029 <div class="quote-msg"> <div class="quote-author">Quote:</div> <p> 1) In an ideal world, each "xmpp node" would be 'dumb'. At any point I'd like to be able to completely wipe the node and rebuild it from scratch. I'm having a hard time seeing how to do this when MNesia is used in the back-end for clustering, since it seems to require some by-hand steps for the cluster setup. </p></div> <p>One simple solution, but I am not sure it will work (so you better try it first) is to setup the nodes of the cluster. Then stop all ejabberd nodes and backup the Mnesia spool dirs.</p> <p>Each time you want to wipe the databases, simply stop all ejabberd nodes, recover the initial spool dirs, and start the nodes again.</p> <div class="quote-msg"> <div class="quote-author">Quote:</div> <p> 2) When you use a clustered setup of XMPP nodes, how do you balance the connections? Do you use simple DNS round robin or do you need a real load balancer in front? </p></div> <p>You have to tell the clients what node to connect. DNS round robin is enough for that task, right?</p> <p>Related tutorial: <a href="http://www.ejabberd.im/cisco-slb" title="http://www.ejabberd.im/cisco-slb">http://www.ejabberd.im/cisco-slb</a></p> <div class="quote-msg"> <div class="quote-author">Quote:</div> <p> 3) We ran into some MNesia resource limits when we hit about 75k connections split across two servers... when we pull the authentication out of the database and put it into an LDAP cluster, should that improve? </p></div> <p>Sorry, I don't know.</p> <p>If you don't get more detailed answers in this forum, you can also ask in the ejabberd mailing list, or contact for support from ProcessOne (like <noindex><a href="http://www.process-one.net/en/company/contact/" rel="nofollow" >Online contact form</a></noindex>).</p> Tue, 15 Dec 2009 11:01:49 +0000 mfoss comment 55029 at https://www.ejabberd.im