Hi Mark,
thanks for the thorough inspection of my results. I'll try to make sure
that I won't include now known false positives in future reports.
Some few remarks inline.
On 04.08.2013 18:14, Mark Thomas wrote:
> On 03/08/2013 12:06, Rainer Jung wrote:
>> Overview:
>>
>> - one crash in tcnative during unit tests
>> - MBeans indicate common loader is gone (is that expected?
>> At least the common.loader entry still exists in catalina.properties)
> Expected as StandardClassLoader has been removed. That decision is
> reversible of there is a use case for having the class loader visible in
> JMX.
So the common loader still exists but is a URLClassLoader? Otherwise I
guess we should remove to common.loader property in catalina.properties?
>> ! Unit tests few failures and errors:
> These need looking at. Note: Many of the tests are timing sensitive and
> do not fail consistently. Making the tests less sensitive to timing is
> an ongoing task.
.. and my build and test system is especially slow (sometimes helpful to
detect race conditions).
The APR crash is probably PR 51813. I added the crash info to BZ after
Chris pointed me to it.
>> - Catalina:j2eeType=WebModule,name=//localhost attribute
>> ignoreAnnotations switched from "true" to "false",
> Need to look in to that.
I double checked, it is really like that.
>> - platform MBean java.lang:type=Threading shows 4 more (daemon)
>> threads
> OK.
Here are the details, which look OK to me, except for Thread-4:
Directly after start:
only for TC 7
- ajp-bio-8009-AsyncTimeout
- http-bio-8080-AsyncTimeout
- Low Memory Detector (JVM internal)
only for TC 8a
- ajp-nio-8009-ClientPoller-0
- ajp-nio-8009-ClientPoller-1
- http-nio-8080-ClientPoller-0
- http-nio-8080-ClientPoller-1
- NioBlockingSelector.BlockPoller-1
- NioBlockingSelector.BlockPoller-2
- Service Thread (JVM internal)
- Thread-4
and after the first request was handled TC 7 also shows
"SnakeWebSocketServlet Timer".
The "Thread-4" has stack
java.lang.Thread.State: RUNNABLE
at sun.nio.ch.SolarisEventPort.portGet(Native Method)
at sun.nio.ch.SolarisEventPort.access$400(SolarisEventPort.java:38)
at
sun.nio.ch.SolarisEventPort$EventHandlerTask.run(SolarisEventPort.java:167)
at java.lang.Thread.run(Thread.java:724)
and gets started between NioBlockingSelector.BlockPoller-2 and
ContainerBackgroundProcessor[StandardEngine[Catalina]]. Not sure whether
it is started by the JVM under the hood, but if we create it ourselves
we should probably give it a name.
>> - platform MBean java.lang:type=OperatingSystem show 88 open file
>> descriptors instead of 61
> Worth just checking what those files are.
I checked again and those are:
- multiple times poll devices and FIFOs (probably due to NIO)
- some additional TC jars (websocket, spdy, jni, storeconfig)
- some addition JDK jars (JDK 7 vs. 6)
So it looks OK to me.
What I noticed: the global web.xml directly after start has 9 FDs open
for it (TC 7) resp. 5 (TC 8). After the first GC those are gone, so
probably only a finalizer thing and not problematic.
>> - org.apache.tomcat.util.net.TestCustomSsl
>> - (old) BIO twice "Exception getting SSL attributes"
>> in org.apache.coyote.http11.Http11Processor actionInternal
>> exception is:
>> javax.net.ssl.SSLProtocolException: handshake alert: no_certificate
>> - (old) NIO twice "Exception re-negotiating SSL connection"
>> in org.apache.coyote.http11.Http11NioProcessor actionInternal
> Need to check.
Note that those also exist for TC 7 at least since 7.0.11.
>> - tribes: several warnings (41), numbers up by 22 from 7.0.42.
>> Mostly same types of messages.
> Probably OK but should check.
The types of messages are the same as for TC 7.
Regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
