https://issues.apache.org/bugzilla/show_bug.cgi?id=55171
Bug ID: 55171
Summary: [jsr 356] All server threads become blocked after some
websocket testing
Product: Tomcat 8
Version: trunk
Hardware: PC
Status: NEW
Severity: critical
Priority: P2
Component: Catalina
Assignee: dev@tomcat.apache.org
Reporter: s.bos...@gmail.com
Created attachment 30512
--> https://issues.apache.org/bugzilla/attachment.cgi?id=30512&action=edit
full thread stack dump
I have a websocket test which performs some stress testing over an echo example
server running on Tomcat 8.
My test is creating up to 200 websocket connections and perform some message
sending/receiving. At some point of time after some test restarts, the server
stops to respond. I have restarted the test with a simple single socket
connection and the server was still not responding. I checked the threads
stacks at the server side and it seems all of the Tomcat worker therads were
blocked.
Attaching the full stack trace of all threads.
In the server's logs file I can see only several exceptions like this:
Jul 01, 2013 11:34:52 AM
org.apache.tomcat.websocket.server.WsHttpUpgradeHandler destroy
SEVERE: Failed to close WebConnection while destroying the WebSocket
HttpUpgradeHandler
java.io.IOException: Broken pipe
at sun.nio.ch.FileDispatcherImpl.write0(Native Method)
at sun.nio.ch.SocketDispatcher.write(SocketDispatcher.java:50)
at sun.nio.ch.IOUtil.writeFromNativeBuffer(IOUtil.java:94)
at sun.nio.ch.IOUtil.write(IOUtil.java:51)
at sun.nio.ch.SocketChannelImpl.write(SocketChannelImpl.java:450)
at
org.apache.tomcat.util.net.SecureNioChannel.flush(SecureNioChannel.java:135)
at
org.apache.tomcat.util.net.SecureNioChannel.close(SecureNioChannel.java:385)
at
org.apache.coyote.http11.upgrade.NioServletInputStream.doClose(NioServletInputStream.java:107)
at
org.apache.coyote.http11.upgrade.AbstractServletInputStream.close(AbstractServletInputStream.java:128)
at
org.apache.coyote.http11.upgrade.AbstractProcessor.close(AbstractProcessor.java:59)
at
org.apache.tomcat.websocket.server.WsHttpUpgradeHandler.destroy(WsHttpUpgradeHandler.java:137)
at
org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:684)
at
org.apache.coyote.http11.Http11NioProtocol$Http11ConnectionHandler.process(Http11NioProtocol.java:223)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1584)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.run(NioEndpoint.java:1543)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:789)
Note: My test setup includes proxy and loadbalancer in the middle. In some
cases the loadbalancer closes websocket connections which is seen as connection
reset by the server and the client.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
