Our CAS servers experienced a failure that wasn't CAS's fault.

One of our LDAP servers ran out of memory but the load balancer kept sending it 
requests from our CAS servers. The CAS server seem to just wait patiently while 
the LDAP server never replied. This seemed to have created a situation where 
the connection to the JPA ticket registry was left open for a long time. The 
end result was login requests queuing up and CAS used up all the connections on 
the Database.

>From there things went downhill fast.

We removed the errant LDAP server from the configuration and restarted CAS, 
then everything went back to normal. It took us about an hour to figure out the 
problem.

Anyway I reviewed the CAS log files and the only errors I could find were 
related to the DB refusing connections.

Does anyone have a suggestion on how I could get CAS to page me if/when this 
situation occurs again. I assume I'll need to go into the code and add the 
necessary logic. I wonder if someone could point to the correct place to add 
this feature.

Thanks,

Bryan

-- 
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

Reply via email to