Right, but the heap doesn't lie :) On Tue, Jun 14, 2016 at 5:04 PM, Romain Manni-Bucau <[email protected]> wrote:
> Not known and quite unlikely since it is the tomee/openejb only logger so > should be very few mem. > > What is your logging config? > > Can you reproduce it? Have some instance in prod for weeks without issues > but these ones are not in 1000TPS family so can have missed sthg. > Le 14 juin 2016 22:56, "sgjava" <[email protected]> a écrit : > > > I'm running into a situation where after about a week of runtime the heap > > gets chewed up. I ran MAT on a heap dump and get: > > > > The class "org.apache.openejb.util.Logger", loaded by > > "java.net.URLClassLoader @ 0x83332c88", occupies 1,581,490,368 (93.31%) > > bytes. The memory is accumulated in one instance of > > "java.util.concurrent.ConcurrentHashMap$Node[]" loaded by "<system class > > loader>". > > > > Now I'm using logback similar to this > > > > > http://hwellmann.blogspot.com/2012/11/logging-with-slf4j-and-logback-in.html > > > > Is this a known issue? > > > > > > > > -- > > View this message in context: > > > http://tomee-openejb.979440.n4.nabble.com/TomEE-7-logger-heap-leak-tp4678876.html > > Sent from the TomEE Users mailing list archive at Nabble.com. > > > -- Steven P. Goldsmith
