When diameter requests get timed out, it tries to destroy the thread group,
(concurrentFactory.getThreadGroup().destroy(); in
org.jdiameter.client.impl.controller.PeerTableImpl)
This line Throws java.lang.IllegalThreadStateException and thread group is
stile alive. So there are lot of unused but not destroyed threads.
As a result we get many
java.lang.OutOfMemoryError: unable to create new native thread
errors after some time.
1.5.1 Final on Linux
Part of the error we get as follows ...
0-23@03:32:51 DEBUG (TCPTransportClient.java:139) - Stopping transport
10-23@03:32:51 DEBUG (TCPTransportClient.java:120) - Transport exception
java.nio.channels.AsynchronousCloseException
at java.nio.channels.spi.AbstractInterruptibleChannel.end(AbstractInterruptibleChannel.java:185)
at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:263)
at org.jdiameter.client.impl.transport.tcp.TCPTransportClient.run(TCPTransportClient.java:106)
at java.lang.Thread.run(Thread.java:619)
10-23@03:32:51 DEBUG (PeerImpl.java:97) - Connection from
aaa://172.25.145.154:19005 is close
10-23@03:32:51 DEBUG (TCPTransportClient.java:134) - Read thread is stopped
10-23@03:32:51 INFO (PeerFSMImpl.java:136) - Event Queue: []
10-23@03:32:51 DEBUG (TCPTransportClient.java:146) - Transport is stopped
10-23@03:32:51 DEBUG (PeerImpl.java:528) - Disconnected from peer
aaa://172.25.145.154:19005
10-23@03:32:51 DEBUG (PeerFSMImpl.java:210) - aaa://172.25.145.154:19005
fsm swith state STOPPING -> DOWN
10-23@03:32:51 DEBUG (PeerFSMImpl.java:155) - Process event
Event{name:START_EVENT, key:aaa://172.25.145.154:19005, object:null}
10-23@03:32:51 DEBUG (PeerFSMImpl.java:144) - Peer fsm stopped
10-23@03:32:51 WARN (PeerTableImpl.java:240) - Can not stop executor
10-23@03:32:51 DEBUG (PeerFSMImpl.java:144) - Peer fsm stopped
10-23@03:32:51 ERROR (JdiameterMessagingServiceImpl.java:257) - Can not
send message for Session-Id [bbe11c16-16bc-48bb-9fbe-34b967ad10c8] to
server,reason: Timeout occured
Original issue reported on code.google.com by manojsam...@gmail.com on 26 Oct 2012 at 5:19
Original issue reported on code.google.com by
manojsam...@gmail.com
on 26 Oct 2012 at 5:19