We're seeing an issue with the ehCache extension (...
# lucee
d
We're seeing an issue with the ehCache extension (using a manual replication configuration) where very occasssionally a node will miss it's instructions to expire the cache. It seems to be very intermittent but I'm trying to track down the root cause. I want to enable debug logging so I can try to at least determine what happened when we see an instance where a node has failed to update the cache. I should be able to enable this debugging by setting the log4j/log4j2
net.sf.ehcache.distribution.RMICachePeer
logger to
debug
. I should know this, but how would I set the logging properties so the extension will pick it up? The docs indicate that you should be able to put a log4j.properties file in the
/lib
folder of Tomcat. Does that still hold true? I assume we'd actually need to use the
log4j2.xml
now that Lucee is running log4j2.
My feeling is I'm probably running into a race condition in ehCache, similar to what's described here: https://stackoverflow.com/questions/35975271/ehcache-dirty-updates-while-replication So there might not be a resolution, without changing the distributed caching mechanism.
z
Good question, Ive also struggled with getting the logging overrides going
d
Thanks. I'll play around with it some tomorrow and post my question to dev.lucee.org too.