I usually look for lines that start with httpWorkerThread, see what they
are waiting on, then check those threads. They mostly show:
> - waiting on <0x108c4f68> (a com.sun.enterprise.web.connector.grizzly.LinkedListPipeline)
I think that is normal. I did notice the selector threads have some
locks, but this could be normal too:
> "SelectorThread-80" prio=6 tid=0x317adc00 nid=0x4b0 runnable [0x357cf000..0x357cfa98]
> java.lang.Thread.State: RUNNABLE
> at sun.nio.ch.WindowsSelectorImpl$SubSelector.poll0(Native Method)
> at sun.nio.ch.WindowsSelectorImpl$SubSelector.poll(WindowsSelectorImpl.java:274)
> at sun.nio.ch.WindowsSelectorImpl$SubSelector.access$400(WindowsSelectorImpl.java:256)
> at sun.nio.ch.WindowsSelectorImpl.doSelect(WindowsSelectorImpl.java:137)
> at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:69)
> - locked <0x108c65f0> (a sun.nio.ch.Util$1)
> - locked <0x108c65e0> (a java.util.Collections$UnmodifiableSet)
> - locked <0x108c6568> (a sun.nio.ch.WindowsSelectorImpl)
> at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:80)
> at com.sun.enterprise.web.connector.grizzly.SelectorThread.doSelect(SelectorThread.java:1337)
> at com.sun.enterprise.web.connector.grizzly.SelectorThread.startListener(SelectorThread.java:1284)
> - locked <0x108c6618> (a [Ljava.lang.Object;)
> at com.sun.enterprise.web.connector.grizzly.SelectorThread.startEndpoint(SelectorThread.java:1247)
> at com.sun.enterprise.web.connector.grizzly.SelectorThread.run(SelectorThread.java:1223)
Beyond looking for things that look like my own thread dumps (http
threads waiting on resource manager, resource manager waiting on JDBC
driver, jdbc driver stuck), I don't know what the rest of this stuff means.
I know Sun can only help non-paying-customers on best effort basis.
Hopefully all of these people having lockups is serious enough for them
to help all of them and fix GlassFish if it is a GlassFish problem.
You're not the first I've seen getting lockups that appear unrelated to
JDBC. I've talked with 2-3 people already who are planning on ditching
GlassFish because of unexplained lockups that they don't get when using
Tomcat, and they are unwilling to pay $4500 to get help when they are
just evaluating GlassFish to see if they want to switch over to it.
Thanks,
Ryan
glassfish_at_javadesktop.org wrote:
> Ryan,
>
> 6 and a half hours this time,
> I've attached a JStack.. I forgot to run the other stack trace.
>
> I changed the way fastpath works, but still get the same problem. I was really hoping it was going to be the jdbc error...
>
> Do you know how to read JStack files? I'm just about to google it, but if anyone has a recommendation on how to read them that will be great...
>
> I look forward to your reply..
> [Message sent by forum member 'adamburrett' (adamburrett)]
>
> http://forums.java.net/jive/thread.jspa?messageID=278931
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>
>
>