Mailing-List: contact issues-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: issues@commons.apache.org
Date: Thu, 8 Nov 2012 21:40:12 +0000 (UTC)
From: "sam marshall (JIRA)" <jira@apache.org>
To: issues@commons.apache.org
Message-ID: <391857349.88968.1352410812739.JavaMail.jiratomcat@arcas>
In-Reply-To: <26892214.88838.1352409132601.JavaMail.jiratomcat@arcas>
Subject: [jira] [Commented] (DAEMON-266) OS X 10.8: jsvc appears to cause
 hang on Toolkit.createDefaultToolkit
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/DAEMON-266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13493517#comment-13493517 ] 

sam marshall commented on DAEMON-266:
-------------------------------------

Here's the stack trace:

{code}
"!!! Problem thread !!!!" prio=5 tid=7f9fba936000 nid=0x10b970000 runnable [10b96e000]
   java.lang.Thread.State: RUNNABLE
	at apple.awt.CToolkit.getContentScaleFactor(Native Method)
	at apple.awt.CToolkit.<init>(CToolkit.java:136)
	- locked <7aad423d8> (a java.lang.Object)
	at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
	at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
	at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
	at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
	at java.lang.Class.newInstance0(Class.java:355)
	at java.lang.Class.newInstance(Class.java:308)
	at java.awt.Toolkit$2.run(Toolkit.java:859)
	at java.security.AccessController.doPrivileged(Native Method)
	at java.awt.Toolkit.getDefaultToolkit(Toolkit.java:826)
	- locked <6fb73d800> (a java.lang.Class for java.awt.Toolkit)
	at Test$1.run(Test.java:22)

   Locked ownable synchronizers:
	- None
{code}

Note: I think this problem may be caused because it's using the wrong java-home. On Mac, the correct java-home can be found by running /usr/libexec/java_home. Using this value, I tried adding this argument to jsvc (this is the same one I used when building it):

{code}
-java-home /Library/Java/JavaVirtualMachines/jdk1.7.0_09.jdk/Contents/Home
{code}

However I then get the error:

{code}
Cannot find any VM in Java Home /Library/Java/JavaVirtualMachines/jdk1.7.0_09.jdk/Contents/Home
Cannot locate JVM library file
Service exit with a return value of 1
{code}

Without doing that, I modified the test script a bit to print out java.home:

{code}
/System/Library/Java/JavaVirtualMachines/1.6.0.jdk/Contents/Home
{code}

I don't think that old Apple 1.6 JDK is really supposed to be used in the current OS X versions (it's not even always installed and is only there to support 'legacy' bundled desktop .app files), so this is probably the root of the problem; jsvc ought to be using the new Java 7 JDK, same that gets used if you type 'java -version' at the commandline. Just a guess though - I'm not an expert in this.

Note: That also might mean that this problem can't be reproduced, or reproduces differently, if you've never run a legacy Java app on the 10.8 Mac. (A legacy Java app is anything that prompts the 'do you want to install java? this will take bloody ages' dialog from Apple.)
                
> OS X 10.8: jsvc appears to cause hang on Toolkit.createDefaultToolkit
> ---------------------------------------------------------------------
>
>                 Key: DAEMON-266
>                 URL: https://issues.apache.org/jira/browse/DAEMON-266
>             Project: Commons Daemon
>          Issue Type: Bug
>          Components: Jsvc
>    Affects Versions: 1.0.10
>         Environment: Max OS X 10.8.2
> (64-bit)
> $ java -version
> java version "1.7.0_09"
> Java(TM) SE Runtime Environment (build 1.7.0_09-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 23.5-b02, mixed mode)
>            Reporter: sam marshall
>         Attachments: jsvc-bug.tar.gz
>
>
> When using jsvc to launch a service (such as Tomcat) on this platform, any use of the Toolkit.createDefaultToolkit causes the current thread to hang indefinitely in native code.
> This does not happen when running the code directly using the normal 'java' runtime.
> I have a gz file containing all files necessary to reproduce this on my machine - will add a comment and attach this file if it lets me, or figure out somewhere to put it.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira