Hello,
I have some more questions about EhCache, especially with respect to a jvm
under duress.
When we get our OutOfMemoryExceptions, a particular peer has a hard time
responding to the other nodes sending tickets, so those nodes also start to
appear sluggish.
What I would like is for the replication to use either async or a timeout value
where if the other peer nodes are under duress, the sending threads don't
block; rather they timeout quickly, or send the tickets a bit more
asynchronously, without the client on the hook. Perhaps the timeout option is
more realistic since the tickets really should make it to the other hosts in
order for the authentication to complete. Unfortunately the load balancer can
remove a host from the front end, but the replication is configured directly
around the back.
I did notice that our heap in production seems to be maxing out at 800MB, so I
am going to increase that to 2G per instance. Any replies on what you are using
for heap settings would be helpful also. We're using Jetty.
I did find a parameter socketTimeoutMillis, to be placed in the
cacheManagerPeerListenerFactory stanza, like this:
<cacheManagerPeerListenerFactory
class="net.sf.ehcache.distribution.RMICacheManagerPeerListenerFactory"
properties="port=9520, socketTimeoutMillis=2000" />
Would this perhaps help us? I'm inclined to add it to our properties, and I've
already done it in test.
I am also reading about a cacheEventListenerFactory that is not included in the
prescribed ehcache-replicated.xml file.
Are people using this listener factory? if so, how is it configured?
Manual versus automatic peer discovery and blocking:
Are there any similar blocking type issues when using multicast? has anyone
else experimented with these items?
I also do have interest in moving the ticket registry (yet again) to memcached,
considering that it would then put the registry outside the JVM heap, so
perhaps the storage/replication would be more reliable for the three node
cluster we are using. Anyone move from ehcache to memcached?
Any feedback on these items would be helpful in our strategy.
Thanks,
-Michael.
--
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
