[seam-issues] [JBoss JIRA] (JBSEAM-4993) Deadlock between SFSB EJB lock and Component.getInstanceFromFactory factoryLock

[Marek Schmidt (JIRA)]

     [ https://issues.jboss.org/browse/JBSEAM-4993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Marek Schmidt updated JBSEAM-4993:
----------------------------------

           Priority: Major  (was: Critical)
        Description: 
The following sample demonstrates a deadlock between SFSB lock and the Component.getInstanceFromFactory factoryLock:

{code}
@Stateful
@Scope(ScopeType.SESSION)
@Name("test")
public class TestAction implements Test
{
   
   public String test()  {        
      Thread.sleep(500);
      Component.getInstance("testString", true);    
      return "test";
   }
   
   @Factory(value="testString", scope=ScopeType.SESSION)
   public String getTestString() {      
      return "testString";
   }

   @Remove
   public void remove() {}
}
{code}


when test.xhtml contains:

{code}
<h:outputText value="#{test.test()} " />
{code}

and othertest.xhtml contains:
{code}
<h:outputText value="#{testString} " />
{code}

If the othertest.seam is requested just after (<500ms after) the test.seam with the same session cookie.

The problem seems to be that the #{test.test()} first locks the TestAction SFSB lock and then the Seam factory lock in the Component.getInstance call, while #{testString} locks the factoryLock first.

  was:
The following sample demonstrates a deadlock between SFSB lock and the Component.getInstanceFromFactory factoryLock:

{code}
@Stateful
@Scope(ScopeType.SESSION)
@Name("test")
public class TestAction implements Test
{
   
   public String test()  {        
      Thread.sleep(500);
      Component.getInstance("foo", true);    
      return "test";
   }
   
   @Factory(value="testString", scope=ScopeType.EVENT)
   public String getTestString() {      
      return "testString";
   }

   @Remove
   public void remove() {}
}
{code}


{code}
@Name("testProducer")
public class TestProducer {
   @Factory(scope=ScopeType.EVENT)
   public String getFoo() {
      return "foo";
   }
}
{code}

when test.xhtml contains:

{code}
<h:outputText value="#{test.test()} " />
{code}

and othertest.xhtml contains:
{code}
<h:outputText value="#{testString} " />
{code}

If the othertest.seam is requested just after (<500ms after) the test.seam with the same session cookie.

The problem seems to be that the #{test.test()} first locks the TestAction SFSB lock and then the Seam factoryLock in the Component.getInstance call, while #{testString} locks the factoryLock first.

    Forum Reference: https://community.jboss.org/message/741120#741120  (was: https://community.jboss.org/message/741120#741120)

    
> Deadlock between SFSB EJB lock and Component.getInstanceFromFactory factoryLock
> -------------------------------------------------------------------------------
>
>                 Key: JBSEAM-4993
>                 URL: https://issues.jboss.org/browse/JBSEAM-4993
>             Project: Seam 2
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 2.3.0.BETA2
>         Environment: JBoss AS 7.1.2.Final
>            Reporter: Marek Schmidt
>            Assignee: Marek Novotny
>             Fix For: 2.3.0.CR1
>
>         Attachments: jstack.txt, registration.tar.bz2, seam-registration.ear, server.log.bz2, test.py
>
>
> The following sample demonstrates a deadlock between SFSB lock and the Component.getInstanceFromFactory factoryLock:
> {code}
> @Stateful
> @Scope(ScopeType.SESSION)
> @Name("test")
> public class TestAction implements Test
> {
>    
>    public String test()  {        
>       Thread.sleep(500);
>       Component.getInstance("testString", true);    
>       return "test";
>    }
>    
>    @Factory(value="testString", scope=ScopeType.SESSION)
>    public String getTestString() {      
>       return "testString";
>    }
>    @Remove
>    public void remove() {}
> }
> {code}
> when test.xhtml contains:
> {code}
> <h:outputText value="#{test.test()} " />
> {code}
> and othertest.xhtml contains:
> {code}
> <h:outputText value="#{testString} " />
> {code}
> If the othertest.seam is requested just after (<500ms after) the test.seam with the same session cookie.
> The problem seems to be that the #{test.test()} first locks the TestAction SFSB lock and then the Seam factory lock in the Component.getInstance call, while #{testString} locks the factoryLock first.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira