[GLASSFISH-18238] promotion frame url should not ref. www.sun.com Created: 23/Jan/12  Updated: 24/Jan/12  Resolved: 24/Jan/12

Status: Resolved
Project: glassfish
Component/s: admin_gui
Affects Version/s: 3.1.2_b18
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Critical
Reporter: Anissa Lam Assignee: Anissa Lam
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Tags: 3_1_2-review

 Description   

GUI code should avoid using www.sun.com.
Searching through all GUI code for www.sun.com , shows the following:

<faces-config xmlns="*http://java.sun.com/xml/ns/javaee*"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="*http://java.sun.com/xml/ns/javaee*
*http://java.sun.com/xml/ns/javaee/web-facesconfig_1_2.xsd*"
version="1.2">

<web-app xmlns="*http://java.sun.com/xml/ns/javaee*"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="3.0"
xsi:schemaLocation="*http://java.sun.com/xml/ns/javaee*
*http://java.sun.com/xml/ns/javaee/web-app_3_0.xsd*">

promotionLink=www.sun.com/software/products/appsrvr/promotions/v3.html

<!DOCTYPE sun-web-app PUBLIC '-//Sun Microsystems, Inc.//DTD Sun ONE
Application Server 7.0 Servlet 2.3//EN'
'*http://www.sun.com/software/sunone/appserver/dtds/sun-web-app_2_3-0.dtd*'>

Will change the promotion link to http://java.sun.com/glassfish/productmsg.html
dtds form www.sun.com should be fine as thsee will be redirected, otherwise all existing J2EE aps will break.



 Comments   
Comment by Anissa Lam [ 23/Jan/12 ]
  • What is the impact on the customer of the bug?
    If the redirects stop working for the promotion link, the home page (commonTask page) will be messed up.
  • What is the cost/risk of fixing the bug?
    easy change, just change the Strings.properties file.
  • Is there an impact on documentation or message strings?
    No.
  • Which tests should QA (re)run to verify the fix did not destabilize GlassFish?
    Ensure the common task page comes up fine, with and without secure admin enabled.
  • Which is the targeted build of 3.1.2 for this fix?
    b19
Comment by scatari [ 24/Jan/12 ]

Must fix for 3.1.2.

Comment by Anissa Lam [ 24/Jan/12 ]

Fix checked into both GF and Oracle GF code base.
Log Message:
------------
GLASSFISH-18238. Change promotion link for not going to www.sun.com
Revisions:
----------
52255
Modified Paths:
---------------
branches/3.1.2/admingui/community-theme/src/main/resources/org/glassfish/admingui/community-theme/Strings.properties

==========
Oracle branding in internal workspace

Revision: 3634
Author : anissa.lam@oracle.com
Date : Jan 23, 2012 9:49:12 PM
GLASSFISH-18238. Change promotion link for not going to www.sun.com





[GLASSFISH-18229] Broken links in index.html and about.html Created: 19/Jan/12  Updated: 25/Jan/12  Resolved: 25/Jan/12

Status: Resolved
Project: glassfish
Component/s: docs
Affects Version/s: 3.1.2_b01
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Critical
Reporter: scatari Assignee: Paul Davies
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

This is not OS dependent.


Tags: 3_1_2-approved

 Description   

Following changes should be addressed in index.html and about.html

This is a must fix for 3.1.2, hence marking as pre-approved for integration.



 Comments   
Comment by Paul Davies [ 21/Jan/12 ]

The addition of the Stay Connected bar is a request for enhancement. Adding such a bar in the same style as in the Administration Console entails the addition of several icons to these pages, which is a change that is too risky to do this late in the schedule.

Only the links that are currently being re-directed must be fixed for this release.

Some links in the features.html page still refer to 3.1.1 information. As part of the fix for this issue, these links must be updated.

Comment by Alexis MP [ 22/Jan/12 ]

If adding images to a static page is hard, let's add text links in the following paragraph (before the "Join the GlassFish community" paragraph) :

<h2>Stay Connected</h2>
<p>
Consider watching a few <a href="http://youtube.com/user/GlassFishVideos">GlassFish videos</a>, make sure you get the latest <a href="http://glassfish.java.net/public/downloadsindex.html">GlassFish Downloads</a>, stay connected with your favorite social network (<a href="http://twitter.com/glassfish">Twitter</a>, <a href="http://www.linkedin.com/groups?home=&gid=106819">LinkedIn</a>, <a href="http://facebook.com/glassfish">Facebook</a>), and stay in touch with daily news from <a href="http://blogs.oracle.com/theaquarium">TheAquarium blog</a>!
</p>

Comment by Alexis MP [ 22/Jan/12 ]

Looking further at index.html :

In the "Install and update additional software components" :

  • OSGi HTTP Service => replace with "GlassFish OSGi Feature Pack"
  • Generic Resource Adapter for JMS => OK
  • OSGi Administration Console => OK
    also add:
  • Additional Examples and Documentation (Jersey, ...)

Also in that same paragraph about the web profile and what can be gained from moving to the full platform:

  • Enterprise Java Beans => replace with "Full support for Enterprise JavaBeans (EJB)"
  • Metro => replace with "Metro Enterprise SOAP Web Service Stack"
  • Jersey => Jersey is built in GlassFish, including in the web profile. While JAX-RS is indeed not in the web profile, this bullet is ambiguous and should be replaced with "JMS OpenMQ", a more obvious value-add of moving to the full platform

These changes should be also made to docs/about.html.

Comment by Paul Davies [ 25/Jan/12 ]

Fixes to only the links that will break and to old information committed in revision 52253.

All of the changes that were suggested in the description and in further comments are requests for enhancement that are not really must fix for this release.





[GLASSFISH-18228] The OSGi Admin Console seems to randomly swap between port 8080 and 4848 Created: 19/Jan/12  Updated: 19/Mar/12  Resolved: 24/Jan/12

Status: Resolved
Project: glassfish
Component/s: OSGi
Affects Version/s: None
Fix Version/s: 3.1.2_b19, 4.0_b21

Type: Bug Priority: Major
Reporter: aaronjwhiteside Assignee: Sanjeeb Sahoo
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Status Whiteboard:

Workaround:
In order to cause webconsole to bind to HTTP Service listen on 8080 (default http port), create a file called
glassfish/domain1/autodeploy/bundles/org.apache.felix.webconsole.internal.servlet.OsgiManager.cfg
with following content:
http.service.filter=VirtualServer=server

Tags: 3_1_2_approved

 Description   

The OSGi Admin Console seems to randomly swap between port 8080 and 4848

When running on 8080 it is accessible from the Server -> OSGi Console link under http://localhost:4848/

However when it magically/randomly flips to running on 4848 the Server -> OSGi Console link returns a 404 error.

I haven't done enough investigation to determine what is causing the port flipping. But it seems extremely strange that it happens at all.

Off the top of my head (without any investigation), if two instances of HttpService are registered the ordering would be random and sometimes it gets the correct one and others it gets the wrong one... shrug



 Comments   
Comment by Sanjeeb Sahoo [ 21/Jan/12 ]

Yes, I have seen this issue and the submitter has analysed it correctly, thanks. This is a regression. See GLASSFISH-12359 where we had earlier fixed the issue, but it has resurfaced as we have stoppsed picking cfg files from modules/autostart/.

Comment by Sanjeeb Sahoo [ 21/Jan/12 ]

See workaround.

Comment by Sanjeeb Sahoo [ 24/Jan/12 ]

trunk: svn rev #52257

A appserver/osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi/felixwebconsoleextension/FelixWebConsoleExtensionActivator.java
A appserver/osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi/felixwebconsoleextension/GlassFishBrandingPlugin.java
D appserver/osgi-platforms/felix-webconsole-extension/src/main/resources/META-INF/webconsole.properties
M appserver/osgi-platforms/felix-webconsole-extension/osgi.bundle
M appserver/osgi-platforms/felix-webconsole-extension/pom.xml
D nucleus/osgi-platforms/felix/src/main/resources/glassfish/modules/autostart/org.apache.felix.webconsole.internal.servlet.OsgiManager.cfg

trunk: svn rev #52272

M appserver/packager/felix/src/main/resources/pkg_proto.py

Comment by Sanjeeb Sahoo [ 24/Jan/12 ]
  • What is the impact on the customer of the bug?

Yes, this is a regression. It has been filed by a user and seen by others as well.

  • What is the cost/risk of fixing the bug?

No risk, as the fix is in an addon component.

  • Is there an impact on documentation or message strings?

No.

  • Which tests should QA (re)run to verify the fix did not destabilize GlassFish?

NA

  • Which is the targeted build of 3.1.2 for this fix?
    b19
Comment by Sanjeeb Sahoo [ 24/Jan/12 ]

3.1.2 branch:
Sending osgi-platforms/felix/src/main/resources/glassfish/modules/autostart
Deleting osgi-platforms/felix/src/main/resources/glassfish/modules/autostart/org.apache.felix.webconsole.internal.servlet.OsgiManager.cfg
Sending osgi-platforms/felix-webconsole-extension
Sending osgi-platforms/felix-webconsole-extension/osgi.bundle
Sending osgi-platforms/felix-webconsole-extension/pom.xml
Adding osgi-platforms/felix-webconsole-extension/src/main/java
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi/felixwebconsoleextension
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi/felixwebconsoleextension/FelixWebConsoleExtensionActivator.java
Adding osgi-platforms/felix-webconsole-extension/src/main/java/org/glassfish/osgi/felixwebconsoleextension/GlassFishBrandingPlugin.java
Deleting osgi-platforms/felix-webconsole-extension/src/main/resources/META-INF/webconsole.properties
Sending packager/felix/src/main/resources/pkg_proto.py
Transmitting file data ...
Committed revision 52276.

Comment by aaronjwhiteside [ 14/Mar/12 ]

This isn't actually fixed in 3.1.2 - it is still happening...

Comment by Sanjeeb Sahoo [ 15/Mar/12 ]

Are you sure it didn't work for you? I specifically bind the web console to http service corresponding to port 8080, so I can't understand why it didn't work for you. I tried reproducing and could not. I doubt you are not using correct version of glassfish-osgi-gui.zip. Download version 3.1.2 of this zip and use.

Comment by aaronjwhiteside [ 15/Mar/12 ]

Shouldn't it be bound to the admin listener and not the normal server/application http listener?

We typically restrict access to the admin ports, even though the osgi console is password protected I'd rather not have it be served up at all to clients.

Btw the glassfish-osgi-gui.zip artifact hasn't been pushed out to the maven repositories yet. I know I can use the auto update tool, but that's not how we deploy into production (we pre-package everything into a tar.gz using the maven assembly plugin). So if you could ensure that everything is published correctly I would be grateful.

http://mvnrepository.com/artifact/org.glassfish.packager/glassfish-osgi-gui

I assume the -bx versions are not the official stable/final releases?

Comment by Sanjeeb Sahoo [ 15/Mar/12 ]

You are looking at the wrong artifact. Our build team have renamed the groupId for some good reason. The artifact is actually available at:

http://search.maven.org/#artifactdetails|org.glassfish.main.packager|glassfish-osgi-gui|3.1.2|distribution-base-zip

Whether the console should be available at 8080 or 4848 is a different issue. We have currently made it available at 8080. If you want to change it to 4848, please try adding autodeploy/bundles/org.apache.felix.webconsole.internal.servlet.OsgiManager.cfg with following content:
http.service.filter=VirtualServer=__asadmin

Comment by aaronjwhiteside [ 15/Mar/12 ]

Oh thanks for the info about the groupId rename, much appreciated.

Will the link in the admin console to the osgi console still work if I adding the custom OsgiManager.cfg file?

Comment by Sanjeeb Sahoo [ 15/Mar/12 ]

yes, the cfg file should still work.

Comment by aaronjwhiteside [ 16/Mar/12 ]

I tried putting the org.apache.felix.webconsole.internal.servlet.OsgiManager.cfg file under modules/autostart but it doesn't seem to work there. I don't want to put it under <domain>/autodeploy/bundles as it is a system level configuration detail and not related to the application.

Comment by Sanjeeb Sahoo [ 16/Mar/12 ]

No, you can't put it under modules/autostart. You have to place it under a directory that's monitored by fileinstall and autodeploy/bundles/ is that directory.

Comment by aaronjwhiteside [ 16/Mar/12 ]

OK, do you want me to open another bug to have the OSGi console run on the admin listener (4848) instead of the http-listener (8080)? Or can you reuse this issue?

Comment by Sanjeeb Sahoo [ 16/Mar/12 ]

Pl. open a new enhancement request if you are not able to configure it using cfg file as suggested earlier.

I don't understand your point of using modules/autostart for cfg file. You are not supposed to modify anything inside installation directory. domain_dir is a user controlled area and hence you should use that to do any domain level configuration such as this.

Comment by aaronjwhiteside [ 16/Mar/12 ]

I'll open a new issue, the point was I shouldn't have to be telling the OSGi console to run on the admin port. It should be doing that by default. Running an admin console even if it is OSGi's admin console on the normal application port (8080) is not a good idea, IMHO.

Comment by Sanjeeb Sahoo [ 17/Mar/12 ]

I certainly don't understand why admin port is different from 8080 for an appserver. What advantage it has for an app server which is always behind firewall? Do you understand?

Comment by aaronjwhiteside [ 19/Mar/12 ]

Usually only the admin port is locked behind a firewall... and 8080 is mapped to 80 on the outside world. My point is it requires extra effort to block /osgi/* at the load balancer while allowing everything else through.

If the default were to keep all the administration related functionality restricted to the admin port (4848) then we wouldn't need any extra configuration to block access to /osgi/*.





[GLASSFISH-18224] Not able to deploy to maven.java.net due to invalid signature Created: 19/Jan/12  Updated: 19/Jan/12  Resolved: 19/Jan/12

Status: Resolved
Project: glassfish
Component/s: None
Affects Version/s: 3.1.2_b17
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Major
Reporter: janey Assignee: sirajg
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Tags: 3_1_2-approved

 Description   

An ant task is added during install phase that repackages the war artifact in the module admingui/war and it's causing invalid gpg signature when deploying to maven.java.net repo.
-Invalid Signature: '/org/glassfish/main/admingui/war/3.1.2-b17/war-3.1.2-b17.war.asc' is not a valid signature for 'war-3.1.2-b17.war'.

Artifacts are now signed during verify phase when deploying to maven.java.net repository.

The fix is to change "install" phase to "packaging" phase.



 Comments   
Comment by scatari [ 19/Jan/12 ]

Approved as this is required for deployment to maven.java.net.

Comment by sirajg [ 19/Jan/12 ]

Diffs :

— pom.xml (revision 52204)
+++ pom.xml (working copy)
@@ -62,7 +62,7 @@
<version>2.0.2</version>
<executions>
<execution>

  • <phase>install</phase>
    + <phase>package</phase>
    <configuration>
    <archive>
    <manifestEntries>
    @@ -79,7 +79,7 @@
    <artifactId>maven-antrun-plugin</artifactId>
    <executions>
    <execution>
  • <phase>install</phase>
    + <phase>package</phase>
    <configuration>




[GLASSFISH-18223] JMS config section in Create cluster page does not hide the custom config section once custom Radio button is selected Created: 19/Jan/12  Updated: 19/Jan/12  Resolved: 19/Jan/12

Status: Resolved
Project: glassfish
Component/s: admin_gui
Affects Version/s: 3.1.2_b17
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Critical
Reporter: Anissa Lam Assignee: Jason Lee
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: PNG File data-lost.png     PNG File invalid-config.png    
Tags: 3_1_2-approved

 Description   

In the create new cluster page, the custom jms config section is not hidden if you select custom and then change back to use Default MQ cluster config.
This is very confusing to user, eg,
1. user 'seems' to be allowed to do an Embedded HA cluster in Local mode which should never happen. (attached invalid-config).
2. user wants to have shared DB, and filled in all the required info, but then, the cluster is created using Default mode (attached, user-data-lost).

Once Default mode is selected, the custom jms section needs to be hidden.



 Comments   
Comment by Jason Lee [ 19/Jan/12 ]

Simple fix:

Index: jms-plugin/src/main/resources/configureJmsCluster.js
===================================================================
--- jms-plugin/src/main/resources/configureJmsCluster.js	(revision 52149)
+++ jms-plugin/src/main/resources/configureJmsCluster.js	(working copy)
@@ -64,14 +64,14 @@
 }
     
 function updateJmsPropertySheet(customConfig) {
+    var jmsTypeSheet = 'propertyForm:jmsTypePropertySheet';
+    var jmsPropsheet = 'propertyForm:jmsPropertySheet';
     if (!customConfig) {
         setDisplay(jmsTypeSheet, 'none');
         setDisplay(jmsPropsheet, 'none');
         return;
     }
     
-    var jmsTypeSheet = 'propertyForm:jmsTypePropertySheet';
-    var jmsPropsheet = 'propertyForm:jmsPropertySheet';
     var baseId = jmsPropsheet + ':configureJmsClusterSection';
     var configStoreType = document.getElementById(baseId+':ConfigStoreTypeProp:configStoreType').value;
     var messageStoreType = document.getElementById(baseId+':MessageStoreTypeProp:messageStoreType').value;
Comment by Jason Lee [ 19/Jan/12 ]
  • What is the impact on the customer of the bug?

Data may be lost if the Custom radio button is not selected

  • What is the cost/risk of fixing the bug?

Minimial. Two line JS change (see diff above)

  • Is there an impact on documentation or message strings?

No.

  • Which tests should QA (re)run to verify the fix did not destabilize GlassFish?

ClusterTest in Console devtests

  • Which is the targeted build of 3.1.2 for this fix?

b19

Comment by Jason Lee [ 19/Jan/12 ]

Fix committed.

Comment by Jason Lee [ 19/Jan/12 ]

Fix revision is r52209.





[GLASSFISH-18208] probable ssh connection leak: ssh connection between das and agent machines sometimes dead while running HA test on windows 2008 Created: 18/Jan/12  Updated: 24/Jan/12  Resolved: 24/Jan/12

Status: Resolved
Project: glassfish
Component/s: admin
Affects Version/s: 3.1.2_b17
Fix Version/s: 3.1.2_b19, 4.0

Type: Bug Priority: Critical
Reporter: sonialiu Assignee: Yamini K B
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

windows 2008


Attachments: File jmap.file.after     File jmap.file.before     File jstack.file.after    
Issue Links:
Related
is related to GLASSFISH-16910 create-instance leaks one thread Resolved
Tags: 3_1_2-approved

 Description   

build: v3.1.2 promoted build17
OS: windows 2008
Has this passed before: Yes, with GF 3.1.

While running HA/MQ test on windows 2008, the ssh connection sometimes randomly dead and displayed the following error when staring cluster/instances:

"Command execution failed. There was a problem while connecting to bigapp-x2250-3.us.oracle.com:22:Connection reset
[testng]
[testng] Please verify you have SSH configured correctly on your system with the proper attributes set on node agent2. You may use update-node-ssh to modify these attributes. See the DAS log file for more information.
[testng]
[testng]
[testng] [testng] To complete this operation run the following command locally on host bigapp-x2250-3.us.oracle.com from the GlassFish install location C:/ha/glassfish3:
[testng]
[testng] bin/asadmin start-local-instance --node agent2 --sync normal instance107
[testng]
[testng] instance102: Could not start instance instance102 on node agent3 (ha-qe-2.us.oracle.com).
[testng]
[testng]
[testng]
[testng] This command requires connecting to host ha-qe-2.us.oracle.com using SSH to complete its operation, but it failed to connect:
[testng]
[testng] Command execution failed. There was a problem while connecting to ha-qe-2.us.oracle.com:22:Connection reset
[testng]
[testng] Please verify you have SSH configured correctly on your system with the proper attributes set on node agent3. You may use update-node-ssh to modify these attributes. See the DAS log file for more information.
[testng]
"

I tried manually to do ssh from the das machine jed-asqe-23 to the agent machines ha-qe-2 and bigapp-x2250-3, all failed. See below:

$ ssh ha-qe-2 uname -a
ssh_exchange_identification: Connection closed by remote host

$ ssh bigapp-x2250-3.us.oracle.com uname -a
ssh_exchange_identification: Connection closed by remote host

**After I restarted domain, the ssh connection rebuilt.**

I collected the jmap of the domain process before the ssh connection stopped and after by using the command jmap -histo:live domain-pid > jmap.file (domain-pid is the pid of the domain's process id)
I also collected the jstack of the domain process after the ssh connection dead using command jstack -l domain-pid > jstack.file

The jmap.file.before(the generated jmap file before the ssh connection stopped), jmap.file.after(the generated jmap file after the ssh connection stopped) and jstack.file.after files have been attached to the bug. Please let me know if you want to look at the HA test systems to get more details.



 Comments   
Comment by varunrupela [ 19/Jan/12 ]

Updated summary to indicate a ssh connection leak.
jmap.file.before shows 24 instances of com.trilead.ssh2.Connection
jmap.file.before shows 104 instances of com.trilead.ssh2.Connection
jstack shows ~150 threads in waiting state.

Comment by Joe Di Pol [ 19/Jan/12 ]

An example stack trace from jstack.file.after:

"Thread-293" daemon prio=6 tid=0x0000000022203000 nid=0x1500 runnable [0x000000002542e000]
java.lang.Thread.State: RUNNABLE
at java.net.SocketInputStream.socketRead0(Native Method)
at java.net.SocketInputStream.read(SocketInputStream.java:150)
at java.net.SocketInputStream.read(SocketInputStream.java:121)
at com.trilead.ssh2.crypto.cipher.CipherInputStream.fill_buffer(CipherInputStream.java:41)
at com.trilead.ssh2.crypto.cipher.CipherInputStream.internal_read(CipherInputStream.java:52)
at com.trilead.ssh2.crypto.cipher.CipherInputStream.getBlock(CipherInputStream.java:79)
at com.trilead.ssh2.crypto.cipher.CipherInputStream.read(CipherInputStream.java:108)
at com.trilead.ssh2.transport.TransportConnection.receiveMessage(TransportConnection.java:232)
at com.trilead.ssh2.transport.TransportManager.receiveLoop(TransportManager.java:672)
at com.trilead.ssh2.transport.TransportManager$1.run(TransportManager.java:470)
at java.lang.Thread.run(Thread.java:722)

Comment by Joe Di Pol [ 19/Jan/12 ]

The stack trace is very similar to those in GLASSFISH-16910 which, in theory, was fixed in 3.1.2.

Comment by Yamini K B [ 20/Jan/12 ]

It appears this bug is not just on Windows. Looking at the code and the test logs, ssh connection leak is happening from stop-instance/stop-cluster. I checked the Solaris HA setup as well and the problem is seen there too. I then reproduced the problem on my setup and have verified a fix as well. I will check it in after I get an approval.

Comment by Yamini K B [ 20/Jan/12 ]
  • What is the impact on the customer of the bug?

Problem is reproducible very easily, ssh connections are being leaked (1 per invocation of stop-instance or stop-cluster) This can be serious when the resource limits are reached. Like mentioned in the description, ssh connections will start failing.

  • What is the cost/risk of fixing the bug?

Fix is simple and low risk since it doesn't impact any other module.

  • Is there an impact on documentation or message strings?

No.

  • Which tests should QA (re)run to verify the fix did not destabilize GlassFish?

HA tests (which continuously stop, start clusters)

  • Which is the targeted build of 3.1.2 for this fix?

B19

Comment by Yamini K B [ 24/Jan/12 ]

Fixed the connection leaks in setup-ssh (remote), install-node, uninstall-node, stop-instance
3.1.2: Rev52249
Trunk: Rev52254





[GLASSFISH-18207] [UB]jpa createEMF failure with security manager on Created: 18/Jan/12  Updated: 25/Jan/12  Resolved: 23/Jan/12

Status: Resolved
Project: glassfish
Component/s: docs
Affects Version/s: 3.1.2_b17
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Major
Reporter: sherryshen Assignee: Rebecca Parks
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Red Head Linux2.6.18-128.el5
JDK 1.6.0_30



 Description   

jpa createEMF failure with security manager on

glassfish-3.1.2-b17.zip

2 jpa createEMF tests failed withh security manager on
appserver-sqe/pe/ejb/ejb30/war/jpaJTACreateEMF
appserver-sqe/pe/ejb/ejb30/war/jpaRLCreateEMF

The server.log shows:
java.security.AccessControlException: access denied (java.lang.reflect.ReflectPermission suppressAccessChecks)
at org.eclipse.persistence.internal.sessions.DatabaseSessionImpl.initializeDescriptors(DatabaseSessionImpl.java:535)

The same tests passed with security manager off.



 Comments   
Comment by sherryshen [ 18/Jan/12 ]

As Mitesh suggested, I checked option of
"eclipselink.security.usedoprivileged=true"
with security manager on, i.e.
asadmin create-jvm-options -Djava.security.manager
asadmin create-jvm-options -Declipselink.security.usedoprivileged=true
the jpa createEMF tests passed.

Comment by sherryshen [ 19/Jan/12 ]

As I understand from Mitesh, glassfish container will set
"eclipselink.security.usedoprivileged=true" with injectEMF.
I tried to run injectEMF tests before createEMF tests, e.g
appserver-sqe/pe/ejb/ejb30/war/jpaJTAInjectEMF
appserver-sqe/pe/ejb/ejb30/war/jpaJTACreateEMF
appserver-sqe/pe/ejb/ejb30/war/jpaRLCreateEMF
The previous failures in createEMF tests are gone.

Comment by Mitesh Meswani [ 19/Jan/12 ]

EclipseLink requires system property "eclipselink.security.usedoprivileged=true" to start using doPrivileged blocks. JPADeployer sets this property while deploying an app that uses JavaEE pattern of using EM/EMF (that is either injecting or looking up EM/EMF) The test here is using JavaSE pattern to create EMF (Persistence.createEMF() ). This results in JPADeployer not being involved and hence the property not being set.

Since it is possible that end users can run into this situation, it would be nice to document this (like as done here . Transferring to doc to update user's guide for this.

Comment by Mitesh Meswani [ 19/Jan/12 ]

I believe this should be assigned to June for fixing the doc.

Comment by Mitesh Meswani [ 19/Jan/12 ]

Assigning to Rebecca as per a discussion with her.

Comment by Rebecca Parks [ 19/Jan/12 ]

Please assign a fix version so it will show up in my query.

Comment by Paul Davies [ 21/Jan/12 ]

[UB] - Affects unbundled documentation

Comment by Rebecca Parks [ 23/Jan/12 ]

I wasn't sure whether to put this in the persistence chapter or the security chapter. This is what I did.

In the "Configuring the Java Persistence Provider" chapter of the Application Development Guide, in the "Using Oracle TopLink in GlassFish Server" section, added:

If the security manager is enabled and you are using the Java Persistence API outside the EJB container (in Java SE mode), see "Enabling and Disabling the Security Manager."

In the "Securing Applications" chapter of the Application Development Guide, in the "Enabling and Disabling the Security Manager" section, added:

If the security manager is enabled and you are using the Java Persistence API outside the EJB container (in Java SE mode), the EclipseLink persistence provider requires that you set the eclipselink.security.usedoprivileged JVM option to true as follows:

asadmin create-jvm-options -Declipselink.security.usedoprivileged=true

If the security manager is enabled and you are using the Java Persistence API within the EJB container (in Java EE mode), the EJB container sets this JVM option for you.

Comment by Rebecca Parks [ 25/Jan/12 ]

Per Mitesh's feedback in an email, made the following changes:

In the "Configuring the Java Persistence Provider" chapter of the Application Development Guide, in the "Using Oracle TopLink in GlassFish Server" section:

If the security manager is enabled and you are using the Java Persistence API by calling Persistence.createEMF(), see "Enabling and Disabling the Security Manager."

In the "Securing Applications" chapter of the Application Development Guide, in the "Enabling and Disabling the Security Manager" section:

If the security manager is enabled and you are using the Java Persistence API by calling Persistence.createEMF(), the EclipseLink persistence provider requires that you set the eclipselink.security.usedoprivileged JVM option to true as follows:

asadmin create-jvm-options -Declipselink.security.usedoprivileged=true

If the security manager is enabled and you are using the Java Persistence API by injecting or looking up an entity manager or entity manager factory, the EJB container sets this JVM option for you.





Remove the 'View Log Raw' button from server instance page (GLASSFISH-18186)

[GLASSFISH-18187] Fix OLH that the 'view log raw' feature is not available in 3.1.2 Created: 13/Jan/12  Updated: 25/Jan/12  Resolved: 25/Jan/12

Status: Resolved
Project: glassfish
Component/s: docs
Affects Version/s: 3.1.2_b18
Fix Version/s: 3.1.2_b19

Type: Sub-task Priority: Major
Reporter: Anissa Lam Assignee: Paul Davies
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified


 Comments   
Comment by Paul Davies [ 13/Jan/12 ]

To allow for the possibility of this issue being fixed in the OLH in a future release, I have set the Fix Version to future release.

To ensure that this issue is described in the Release Notes for 3.1.2, I have added the 3_1_2-release-notes tag.

Comment by Paul Davies [ 25/Jan/12 ]

Removed the 3_1_2-release-notes tag. This issue was fixed in revision 52253, so does not need to be added to the Release Notes.

Comment by Paul Davies [ 25/Jan/12 ]

Fix committed in revision 52253.





[GLASSFISH-18181] [OLH] Change text on Domain Logs page Created: 12/Jan/12  Updated: 25/Jan/12  Resolved: 25/Jan/12

Status: Resolved
Project: glassfish
Component/s: docs
Affects Version/s: 3.1.2_b17
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Major
Reporter: lidiam Assignee: Paul Davies
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

ogs-3.1.2-b17.zip


Tags: 312_gui_new, 312_qa

 Description   

Currently the Domain Logs online help page reads:

Note:

The ZIP archive also includes the log file for the domain administration server (DAS), regardless of whether you specify an instance or a cluster.

The above is not correct. DAS logs files are only downloaded with cluster logs but are NOT downloaded with standalone instance logs. I presume this is by design but will ping Anissa for confirmation.



 Comments   
Comment by Paul Davies [ 13/Jan/12 ]

Too late to fix for 3.1.2.

The corrected statement that the DAS log log files are downloaded with cluster logs but not instance logs should also be added to the CLI help for collect-log-files.

Comment by Paul Davies [ 25/Jan/12 ]

Removed the 3_1_2-release-notes tag. This issue was fixed in revision 52253, so does not need to be added to the Release Notes.

Comment by Paul Davies [ 25/Jan/12 ]

Fix committed in revision 52253.





[GLASSFISH-18165] create-password-alias: description of token incorrect Created: 10/Jan/12  Updated: 25/Jan/12  Resolved: 25/Jan/12

Status: Resolved
Project: glassfish
Component/s: docs
Affects Version/s: 3.1.2_b16
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Minor
Reporter: Paul Davies Assignee: Paul Davies
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified


 Description   

The create-password-alias man page states: An alias is a token of the form $

{ALIAS=aliasname-password}

. The token is incorrect and should be: $

{ALIAS=aliasname}

.



 Comments   
Comment by Paul Davies [ 25/Jan/12 ]

Fix committed in revision 52253.





[GLASSFISH-18142] JVM invocation command line is no longer logged Created: 06/Jan/12  Updated: 18/Jan/12  Resolved: 18/Jan/12

Status: Closed
Project: glassfish
Component/s: admin
Affects Version/s: 3.1.2_b09
Fix Version/s: 3.1.2_b19, 4.0

Type: Bug Priority: Critical
Reporter: Joe Di Pol Assignee: naman_mehta
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: Text File 18142.txt    
Issue Links:
Dependency
depends on GLASSFISH-16984 GlassFish v3.1 is chatty like a fish ... Closed
Tags: 3_1_2-approved

 Description   

The JVM invocation command line use to be logged to the DAS log file at server startup. Starting with 3.1.2 B09 it is no longer doing so. This is a loss of valuable diagnostic information.

For example, in B07 after my first start-domain I see the following in the DAS server.log:

Jan 6, 2012 2:22:13 PM com.sun.enterprise.admin.launcher.GFLauncherLogger info
INFO: JVM invocation command line:
/opt/jdk/latest/bin/java
-cp
/export/tmp/glassfish3/glassfish/modules/glassfish.jar
-XX:+UnlockDiagnosticVMOptions
-XX:MaxPermSize=192m
-XX:NewRatio=2
-Xmx512m
-client
-javaagent:/export/tmp/glassfish3/glassfish/lib/monitor/flashlight-agent.jar
-Dosgi.shell.telnet.maxconn=1
-Dfelix.fileinstall.disableConfigSave=false
-Djdbc.drivers=org.apache.derby.jdbc.ClientDriver
-Dfelix.fileinstall.dir=/export/tmp/glassfish3/glassfish/modules/autostart/
-Djavax.net.ssl.keyStore=/export/tmp/glassfish3/glassfish/domains/domain1/config/keystore.jks
-Dosgi.shell.telnet.port=6666
-Djava.security.policy=/export/tmp/glassfish3/glassfish/domains/domain1/config/server.policy
-Djava.awt.headless=true
-Dfelix.fileinstall.log.level=2
-Dfelix.fileinstall.poll=5000
-Dcom.sun.aas.instanceRoot=/export/tmp/glassfish3/glassfish/domains/domain1
-Dosgi.shell.telnet.ip=127.0.0.1
-Dcom.sun.enterprise.config.config_environment_factory_class=com.sun.enterprise.config.serverbeans.AppserverConfigEnvironmentFactory
-Djava.endorsed.dirs=/export/tmp/glassfish3/glassfish/modules/endorsed:/export/tmp/glassfish3/glassfish/lib/endorsed
-Dcom.sun.aas.installRoot=/export/tmp/glassfish3/glassfish
-Dfelix.fileinstall.bundles.startTransient=true
-Djava.ext.dirs=/opt/jdk/latest/lib/ext:/opt/jdk/latest/jre/lib/ext:/export/tmp/glassfish3/glassfish/domains/domain1/lib/ext
-Dfelix.fileinstall.bundles.new.start=true
-Djavax.net.ssl.trustStore=/export/tmp/glassfish3/glassfish/domains/domain1/config/cacerts.jks
-Dcom.sun.enterprise.security.httpsOutboundKeyAlias=s1as
-Djava.security.auth.login.config=/export/tmp/glassfish3/glassfish/domains/domain1/config/login.conf
-DANTLR_USE_DIRECT_CLASS_LOADING=true
-Dgosh.args=--nointeractive
-Djava.library.path=/export/tmp/glassfish3/glassfish/lib:/opt/jdk/jdk1.6.0_23/jre/lib/i386/server:/opt/jdk/jdk1.6.0_23/jre/lib/i386:/opt/jdk/jdk1.6.0_23/lib/i386:/usr/jdk/packages/lib/i386:/lib:/usr/lib
com.sun.enterprise.glassfish.bootstrap.ASMain
-domainname
domain1
-asadmin-args
--host,,,localhost,,,--port,,,4848,,,--secure=false,,,--terse=false,,,--echo=false,,,--interactive=true,,,start-domain,,,--verbose=false,,,--debug=false,,,--domaindir,,,/export/tmp/glassfish3/glassfish/domains,,,domain1
-instancename
server
-verbose
false
-debug
false
-asadmin-classpath
/export/tmp/glassfish3/glassfish/modules/admin-cli.jar
-asadmin-classname
com.sun.enterprise.admin.cli.AsadminMain
-upgrade
false
-type
DAS
-domaindir
/export/tmp/glassfish3/glassfish/domains/domain1
-read-stdin
true
Jan 6, 2012 2:22:14 PM com.sun.enterprise.admin.launcher.GFLauncherLogger info
INFO: Successfully launched in 16 msec.
[#|2012-01-06T14:22:15.681-0800|INFO|glassfish3.1.2|com.sun.enterprise.server.logging.GFFileHandler|_ThreadID=1;_ThreadName=Thread-2;|Running GlassFish Version: GlassFish Server Open Source Edition 3.1.2 (build 7)|#]

[#|2012-01-06T14:22:15.949-0800|INFO|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.services.impl|_ThreadID=14;_ThreadName=Thread-2;|Grizzly Framework 1.9.36 started in: 54ms - bound to [0.0.0.0:3700]|#]

But in 3.1.2 B09 I don't get the JVM invocation line, I just get:

Jan 6, 2012 2:20:57 PM com.sun.enterprise.admin.launcher.GFLauncherLogger info
INFO: Successfully launched in 18 msec.
[#|2012-01-06T14:20:59.062-0800|INFO|glassfish3.1.2|com.sun.enterprise.server.logging.GFFileHandler|_ThreadID=1;_ThreadName=Thread-2;|Running GlassFish Version: GlassFish Server Open Source Edition 3.1.2 (build 9)|#]

[#|2012-01-06T14:20:59.329-0800|INFO|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.services.impl|_ThreadID=11;_ThreadName=Thread-2;|Grizzly Framework 1.9.39 started in: 65ms - bound to [0.0.0.0:4848]|#]

[#|2012-01-06T14:20:59.329-0800|INFO|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.services.impl|_ThreadID=12;_ThreadName=Thread-2;|Grizzly Framework 1.9.39 started in: 88ms - bound to [0.0.0.0:8080]|#]

[#|2012-01-06T14:20:59.330-0800|INFO|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.services.impl|_ThreadID=13;_ThreadName=Thread-2;|Grizzly Framework 1.9.39 started in: 71ms - bound to [0.0.0.0:8181]|#]


 Comments   
Comment by Joe Di Pol [ 07/Jan/12 ]

Increasing priority since this could be considered a regression from 3.1.1

Comment by Tom Mueller [ 08/Jan/12 ]

This message was changed from INFO to FINE based on issue 16984.

Comment by Tom Mueller [ 08/Jan/12 ]

Resolving this as works as designed since this was intentionally changed to be a FINE message based on issue 16984.

Comment by Joe Di Pol [ 09/Jan/12 ]

Ugh. I think the fix for 16984 was a little overzealous. The command line used to run the server is information that would be of interest to an administrator. For example, it confirms whatever JVM arguments an administrator may have configured for the server. I, for one, would like to see this one log message returned.

Comment by Tom Mueller [ 09/Jan/12 ]

Reopening this issue and assigning to Naman. Naman, please discuss with Joe about whether the change introduced with 16984 for this log message should be backed out.

Comment by Anissa Lam [ 09/Jan/12 ]

I vote for Joe. This jvm info is really helpful, and its not that i can change the log level to FINE and can see that again without restarting server.
As of now, even specifying --verbose will not give this JVM info, although the man page says:

--verbose , -v
Specifies whether detailed information about the domain
is displayed in the console window where the subcommand
is run.

I think at least this info should be considered as "detailed information".

Comment by naman_mehta [ 10/Jan/12 ]

So can we do like this?

1. I will club all jvm options on startup and create one string which would be displayed as INFO.
2. All remaining messages would be part of another string which would be displayed as FINE.

Comment by naman_mehta [ 11/Jan/12 ]
  • What is the impact on the customer of the bug?
    This bug shows initial log messages are missing on start up.
  • How likely is it that a customer will see the bug and how serious is the bug?
    It's critical as user is missing jvm options set initially.
  • Is it a regression? Does it meet other bug fix criteria (security, performance, etc.)?
    No
  • What is the cost/risk of fixing the bug?
    When I fixed bug 16984, I moved all initial log messages to FINE level. So now as per the discussion planning to move JVM Opitons details to INFO and remaining message to FINE level.
  • How risky is the fix? How much work is the fix? Is the fix complicated?
    No impact on other modules.
  • Is there an impact on documentation or message strings?
    No.
  • Which tests should QA (re)run to verify the fix did not destabilize GlassFish?
    You can see the messages on start up only.
  • Which is the targeted build of 3.1.2 for this fix?
    3.1.2 b18
Comment by naman_mehta [ 11/Jan/12 ]

Attached file diff for the same...

Comment by naman_mehta [ 18/Jan/12 ]

Fixed for both 3.1.2 and 4.0.





[GLASSFISH-18128] Unlocalized message in the output of the commands in ko and zh_CN locales Created: 05/Jan/12  Updated: 30/Jan/12  Resolved: 18/Jan/12

Status: Closed
Project: glassfish
Component/s: l10n
Affects Version/s: 3.1.2_b16
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Major
Reporter: sunny-gui Assignee: gmurr
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Server OS: Windows 2008 R2 x64 KO w/JDK1.7.0 32bit
Bundle: glassfish-3.1.2-web-b16-windows-ml.exe
Checked Locales: ko and zh_CN


Attachments: HTML File report.all.html     JPEG File unlocalized_strings.jpg    
Issue Links:
Dependency
blocks GLASSFISH-18090 No domain name when execute list-doma... Closed
Tags: 3_1_2-review, l10n

 Description   

Unlocalized message when execute commands in ko and zh_CN locales

To reproduce:
1. After installed GF completely, open terminal window and execute commands, like
asadmin list-domains
asadmin list-backups
asadmin login
asadmin start-database
asadmin stop-database
asadmin create-domain
asadmin delete-domain

— All of the output will be displayed in English.

— In step 1, the output of the commands were included in the file \com\sun\enterprise\admin\cli\optional\LocalStrings_ko.properties(cli-optional-l10n.jar), and those message were localized correctly in the properties file. It seems in Non-EN locale(ko), GF get the message from cli-optional.jar, not cli-optional-l10n.jar

2. Execute command "asadmin stop-domain", the message was localized correctly.

— The localized message was included in the file \com\sun\enterprise\admin\cli\LocalStrings_ko.properties(admin-cli-l10n.jar)

3. Repeat Step 1 and Step 2 in the OEL 5 x86 zh_CN locale with bundle ogs-3.1.2-b16-unix-ml.sh, this issue is reproducible.

Attached screen shot for your reference.



 Comments   
Comment by sunny-gui [ 18/Jan/12 ]

Verified in bundle ogs-3.1.2-web-b18-01_17_2012-ml.zip, this issue is reproducible, attached report file for your reference.

Comment by gmurr [ 18/Jan/12 ]

Fix will be available in b19

Comment by sunny-gui [ 30/Jan/12 ]

Verified and fixed in build 20 nightly, here are details.

Server OS: OEL 6 x64 w/JDK 1.6.0_30 64-bit
Bundle: ogs-3.1.2-b20_01_29_2012-ml.zip
Server Locale: ja_JP.UTF-8





[GLASSFISH-18070] NullPointerException thrown in AppSpecificConnectorClassLoaderUtil Created: 21/Dec/11  Updated: 25/Jan/12  Resolved: 25/Jan/12

Status: Resolved
Project: glassfish
Component/s: jca
Affects Version/s: 3.1.2_b15
Fix Version/s: 3.1.2_b19

Type: Bug Priority: Major
Reporter: David Wang Assignee: skgaju
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Platforms: Windows XP, and Solaris 10 x86
JDK: 1.6.0_24


Attachments: File activemq-rar-5.1.0.rar     Zip Archive EnterpriseApplication.zip     Zip Archive server-logs.zip    
Status Whiteboard:

fixed by internal bug 13590818, this is ported from 3.1.1 bug fix.


 Description   

NullPointerException is thrown by com.sun.appserv.connectors.internal.api.AppSpecificConnectorClassLoaderUtil.detectResourceInRA() when a cluster is started.

This exception occurs when there are two clusters (i.e app and comm).

A EAR file with a single stateless EJB is deployed on the "app" cluster. Furthermore, the stateless EJB has a @Resource annotation for a JMS topic (i.e "jms/TestTopic"), which has not been added to the domain.

A RAR file deployed to the "comm" cluster. In this scenario, the ActiveMQ (5.1.0) RAR file was used. The problem was reproduced with other RARs including custom RARs.

No errors occured during the initial deployment of the EAR and RAR files. When the "app" cluster was restarted, the NullPointerException was thrown.

[#|2011-12-21T12:21:11.421-0800|SEVERE|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.server|_ThreadID=10;_ThreadName=Thread-2;|java.lang.NullPointerException
	at com.sun.appserv.connectors.internal.api.AppSpecificConnectorClassLoaderUtil.detectResourceInRA(AppSpecificConnectorClassLoaderUtil.java:266)
	at com.sun.appserv.connectors.internal.api.AppSpecificConnectorClassLoaderUtil.processDescriptorForRAReferences(AppSpecificConnectorClassLoaderUtil.java:211)
	at com.sun.appserv.connectors.internal.api.AppSpecificConnectorClassLoaderUtil.processDescriptorForRAReferences(AppSpecificConnectorClassLoaderUtil.java:137)
	at com.sun.appserv.connectors.internal.api.AppSpecificConnectorClassLoaderUtil.detectReferredRARs(AppSpecificConnectorClassLoaderUtil.java:118)
	at com.sun.appserv.connectors.internal.api.ConnectorClassLoaderServiceImpl.getConnectorClassLoader(ConnectorClassLoaderServiceImpl.java:111)
	at com.sun.enterprise.v3.server.ClassLoaderHierarchyImpl.getConnectorClassLoader(ClassLoaderHierarchyImpl.java:117)
	at com.sun.enterprise.v3.server.AppLibClassLoaderServiceImpl.getAppLibClassLoader(AppLibClassLoaderServiceImpl.java:94)
	at com.sun.enterprise.v3.server.ClassLoaderHierarchyImpl.getAppLibClassLoader(ClassLoaderHierarchyImpl.java:124)
	at org.glassfish.deployment.common.DeploymentContextImpl.createClassLoader(DeploymentContextImpl.java:213)
	at org.glassfish.deployment.common.DeploymentContextImpl.createApplicationClassLoader(DeploymentContextImpl.java:206)
	at com.sun.enterprise.v3.server.ApplicationLifecycle.deploy(ApplicationLifecycle.java:399)
	at com.sun.enterprise.v3.server.ApplicationLoaderService.processApplication(ApplicationLoaderService.java:375)
	at com.sun.enterprise.v3.server.ApplicationLoaderService.postConstruct(ApplicationLoaderService.java:219)
	at com.sun.hk2.component.AbstractCreatorImpl.inject(AbstractCreatorImpl.java:131)
	at com.sun.hk2.component.ConstructorCreator.initialize(ConstructorCreator.java:91)
	at com.sun.hk2.component.AbstractCreatorImpl.get(AbstractCreatorImpl.java:82)
	at com.sun.hk2.component.SingletonInhabitant.get(SingletonInhabitant.java:67)
	at com.sun.hk2.component.EventPublishingInhabitant.get(EventPublishingInhabitant.java:139)
	at com.sun.hk2.component.AbstractInhabitantImpl.get(AbstractInhabitantImpl.java:78)
	at com.sun.enterprise.v3.server.AppServerStartup.run(AppServerStartup.java:253)
	at com.sun.enterprise.v3.server.AppServerStartup.doStart(AppServerStartup.java:145)
	at com.sun.enterprise.v3.server.AppServerStartup.start(AppServerStartup.java:136)
	at com.sun.enterprise.glassfish.bootstrap.GlassFishImpl.start(GlassFishImpl.java:79)
	at com.sun.enterprise.glassfish.bootstrap.GlassFishDecorator.start(GlassFishDecorator.java:63)
	at com.sun.enterprise.glassfish.bootstrap.osgi.OSGiGlassFishImpl.start(OSGiGlassFishImpl.java:69)
	at com.sun.enterprise.glassfish.bootstrap.GlassFishMain$Launcher.launch(GlassFishMain.java:117)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
	at java.lang.reflect.Method.invoke(Method.java:597)
	at com.sun.enterprise.glassfish.bootstrap.GlassFishMain.main(GlassFishMain.java:97)
	at com.sun.enterprise.glassfish.bootstrap.ASMain.main(ASMain.java:55)

In the detectResourceInRA() method an attempt is made to obtain the missing JMS Resource (lines 221-229). When it is unable to resolve the resource, it enters a while loop (line 261). During the first iteration of the loop, an "appName" of "activemq-rar-5.1.0" was used to obtain an ApplicationInfo instance from the "appRegistry". A null pointer was return from the "appRegistry.get(appName)" method call.

Since the "activemq-rar-5.1.0" RAR was only deployed to the "comm" cluster, the AppSpecificConnectorClassLoaderUtil should not have used it to obtain the missing JMS resource when restarting the "app" cluster.

Attached is a copy of the ActiveMQ 5.1.0 RAR file, a sample EAR project, and glassfish server logs for the domain and both app and comm cluster instances.



 Comments   
Comment by Joe Di Pol [ 25/Jan/12 ]

This looks like it was fixed in 3.1.2 in r52089 via the port of internal bug 12810445





Generated at Mon Dec 05 04:06:18 UTC 2016 using JIRA 6.2.3#6260-sha1:63ef1d6dac3f4f4d7db4c1effd405ba38ccdc558.