[ https://issues.jboss.org/browse/MODCLUSTER-178?page=com.atlassian.jira.pl... ] Roman Jurkov commented on MODCLUSTER-178: ----------------------------------------- Rado, is there a special configuration that i should do to achieve case 3? i've tested it with sticky session true and force to false, and i am getting 503 if node goes down, here is my info: balancer: [1] Name: mycluster Sticky: 1 [JSESSIONID]/[jsessionid] remove: 0 force: 1 Timeout: 0 maxAttempts: 1 node: [2:2],Balancer: mycluster,JVMRoute: MyClusterE1,LBGroup: [MyDomain2],Host: 127.0.0.1,Port: 4453,Type: http,flushpackets: 0,flushwait: 1,ping: 10,smax: 0,ttl: 60,timeout: 0 host: 1 [myClusterAlias2] vhost: 1 node: 2 host: 2 [mycluterAlias1] vhost: 1 node: 2 context: 1 [/cluster] vhost: 1 node: 2 status: 1 my session cookie: id: 6a1f20eda1ec119e7fa168f766f2309bb5f42692.MyClusterD route: MyClusterD as you see node is MyClusterD and route is MyClusterD which died, however there is MyClusterE that serves /cluster context, however mod_cluster doesn't elect that node -- This message was sent by Atlassian JIRA (v6.2.3#6260)