> From what I can see the not is attempting to remove a ticket from it's local > persistence store but failing as it's been already removed by the other node > perhaps, not sure. > ... > zbyeJAktEOLN-login.ulster.ac.uk; nested exception is > javax.persistence.EntityNotFoundException: Unable to find > org.jasig.cas.ticket.TicketGrantingTicketImpl with id > TGT-114064-WcfPfIzfkFOTPMq9MgBDmjsVeh > IFyZBCTwsZURzbyeJAktEOLN-login.ulster.ac.uk
That's possible, but it's also common for the ticket to be deleted by the user initiating logout. In any case the exception above is fairly commonplace. > I've also noticed that there appears to be a lot of tickets in the database > that have expired, but have not been deleted. This should be cause for concern. > Usually shortly after this exception the application server freezes up for > around 30 seconds and service is unavailable on one of the nodes. The timing and apparent correlation with the cleanup process error seems suspicious, but I'm not convinced that one is necessarily related to the other. Can you describe the state of the frozen machine a little more? For example, is it the entire host, the application server, or just the CAS server application? Are there any telling log messages during the frozen period? Have you examined the JVM state during the freeze with a tool such as YourKit? M -- You are currently subscribed to [email protected] as: [email protected] To unsubscribe, change settings or access archives, see http://www.ja-sig.org/wiki/display/JSG/cas-user
