We have several JBoss instances running and use the log4j email appender to monitor the problems which may occur on these server. This is proving to be a real pain in 4.0.4rc1 (it was just fine in 4.0.3) because some code emits stacktraces to System.err (tomcat), and because some messages should probably not be logged at "error" level.
Specifically, in tomcat there is
"
Exception during post-request cleanup.
org.apache.catalina.connector.ClientAbortException
"
I would consider this is message at warning level at most (more likely even debug, as ClaienAbortException cannot be considered specifically exceptional.
Another problem is in jboss cache, it emits "error" messages like
"
2006-02-20 08:51:17,645 ERROR [org.jboss.cache.lock.IdentityLock] write lock for /bla/nla/Bean/bla.bla.Bean#cKLUYAUkrWS2YV could not be acquired after 0 ms. Locks: Read lock owners: [<192.168.0.252:32778>:367]
Write lock owner: null
(caller=Thread[http-192.168.0.252-8080-3,5,jboss], lock info: read owners=[<192.168.0.252:32778>:367] (activeReaders=1, activeWriter=null, waitingReaders=0, waitingWriters=0, waitingUpgrader=0))
"
Again this seems to be more or less normal and should be info or debug level. Now if it could not acquire the lock at all in the allowed wait period, that is an error. Not being able to acquire it immediately seems normal.
Can these be fixed? Do I need to create jira issues for these?
Kind regards,
Joachim
