https://bz.apache.org/bugzilla/show_bug.cgi?id=58646
--- Comment #21 from Alexander Malyshev <anmalys...@mail.ru> ---
Ok, I added some additional logs and here is what we have:
Feb 15, 2016 MSK 05:34:24.157 PM http-nio-135.60.87.68-443-exec-34
org.apache.coyote.http11.Http11NioProtocol
INFO: Starting to process processor with request 500. Procassor source:
connections map.Calling method is:
org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:634)
Feb 15, 2016 MSK 05:34:24.157 PM http-nio-135.60.87.68-443-exec-34
org.apache.coyote.http11.Http11NioProcessor
INFO: Processing GET request to /viewer/wcc/app.swf.
Feb 15, 2016 MSK 05:34:24.157 PM http-nio-135.60.87.68-443-exec-34
com.avaya.acc.base.tomcat.MCPTomcatAccessLogValve
INFO: 135.60.87.91 - - [15/Feb/2016:17:34:23 +0300] "GET /viewer/wcc/app.swf
HTTP/1.1" 200 -
...
Feb 15, 2016 MSK 05:34:24.170 PM http-nio-135.60.87.68-443-exec-34
org.apache.coyote.AbstractProtocol$RecycledProcessors
INFO: Pushing processor with request 500 to RecycledProcessors. Calling method
is:
org.apache.coyote.AbstractProtocol$RecycledProcessors.push(AbstractProtocol.java:906)
Feb 15, 2016 MSK 05:34:24.170 PM http-nio-135.60.87.68-443-exec-34
org.apache.coyote.AbstractProtocol$RecycledProcessors
INFO: Pushing processor with request 500 to RecycledProcessors. Calling method
is:
org.apache.coyote.AbstractProtocol$RecycledProcessors.push(AbstractProtocol.java:906)
As you can see, thread just processes standard static content request and after
request is processed it pushes processor 2 times to the RecycledProcessors
map..
Note that I have incorrect calling method log in last two lines, so it's hard
to say which exact places into
AbstractProtocol.AbstractConnectionHandler.process() was called (it has 5
occurences of release() method call, which puts processor back to
RecycledProcessors map). In case of upgrade request processing it's
theoretically possible (first call in "if (state == SocketState.UPGRADING)
{...}" block and second in the end of process() method if IOExcepton is thrown
after release is called first time). But, having in mind I've not used
websocket at all during this test, I'm not sure how it could call release
twise.
I'll correct last log and try to reproduce again.
--
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
