At our site we have CAS clustering working with an F5 load balancer using sticky sessions. However, if the load balancer detects that a node fails it migrates those sessions to another node where the Tomcat session and ticket cache have already been replicated.
In other words, we use sticky sessions and have uninterrupted service within the cluster. On Tue, 2007-08-07 at 11:05 -0500, Andrew R Feller wrote: > Q: Are there any load balancing requirements for clustering CAS 3 > server? > > > > While following the Clustering CAS walkthrough in the CAS wiki, I > notice it didn’t go in-depth about what options are available. Aside > from using sticky sessions to redirect the user back to the CAS server > they were authenticated against, what other options are available? I > would prefer not to use sticky sessions as we want uninterrupted > service within the cluster. > > > > Thanks, > > > > Andrew R Feller, Analyst > > Subversion Administrator > > University Information Systems > > Louisiana State University > > [EMAIL PROTECTED] > > (office) 225.578.3737 > > > > > _______________________________________________ > Yale CAS mailing list > [email protected] > http://tp.its.yale.edu/mailman/listinfo/cas -- Mike Kennedy Lead Technologist for Research Computing Infrastructure and Security Group [EMAIL PROTECTED] 951.827.4875 _______________________________________________ Yale CAS mailing list [email protected] http://tp.its.yale.edu/mailman/listinfo/cas
