DO NOT REPLY [Bug 45576] New: JAASRealm not working with DigestAuthenticator

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=45576

   Summary: JAASRealm not working with DigestAuthenticator
   Product: Tomcat 6
   Version: 6.0.16
  Platform: PC
OS/Version: Windows XP
Status: NEW
  Severity: normal
  Priority: P2
 Component: Catalina
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


JAASRealm cannot be used with DigestAuthenticator.

JAASRealm only overrides authenticate(String, String), whereas
DigestAuthenticator calls authenticate(String, String, String, String, String,
String, String, String).

This is similar to bug 41407, where CLIENT-CERT authentication is used.


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 45580] New: Unhandled NullPointerException when trying to write to a closed socket.

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=45580

   Summary: Unhandled NullPointerException when trying to write to a
closed socket.
   Product: Tomcat 6
   Version: unspecified
  Platform: PC
OS/Version: Windows XP
Status: NEW
  Severity: normal
  Priority: P2
 Component: Catalina
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


Description of the scenario:

Consider a client using web Messaging service with NIOConnector configured.
Suppose if the client is closed by the time server writes the response to the
socket, it throws a NPE. It appears that there is no check for the null socket
condition. I think the addition of a null check will correct this.

The stack trace of the exception is: 

[ActiveMQMessageConsumer] Exception while processing message:
java.lang.NullPointerException
java.lang.NullPointerException
at
org.apache.coyote.http11.InternalNioOutputBuffer.addToBB(InternalNioOutputBuffer.java:607)
at
org.apache.coyote.http11.InternalNioOutputBuffer.commit(InternalNioOutputBuffer.java:600)
at
org.apache.coyote.http11.Http11NioProcessor.action(Http11NioProcessor.java:1034)
at org.apache.coyote.Response.action(Response.java:183)
at org.apache.coyote.Response.sendHeaders(Response.java:379)
at
org.apache.catalina.connector.OutputBuffer.doFlush(OutputBuffer.java:305)
at
org.apache.catalina.connector.OutputBuffer.close(OutputBuffer.java:273)
at
org.apache.catalina.connector.CoyoteOutputStream.close(CoyoteOutputStream.java:104)


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 44423] invalid duplicate listeners warning

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=44423


Mike Morris <[EMAIL PROTECTED]> changed:

   What|Removed |Added

 Status|RESOLVED|REOPENED
 Resolution|FIXED   |




--- Comment #5 from Mike Morris <[EMAIL PROTECTED]>  2008-08-06 08:45:09 PST ---
I confirm that this is still happening in 6.0.18.  Furthermore there seems to
be a related bug where the contextDestoyed method is ALSO called during the
startup sequence.  Should this be logged as a separate bug?

Logfile trace:

2008/08/06 05:26:11 org.apache.catalina.core.AprLifecycleListener init
INFO: The APR based Apache Tomcat Native library which allows optimal
performance in production environments was not found on the java.library.path:
/home/apps/jdk1.6.0_02/jre/lib/i386/client:/home/apps/jdk1.6.0_02/jre/lib/i386:/home/apps/jdk1.6.0_02/jre/../lib/i386:/usr/java/packages/lib/i386:/lib:/usr/lib
2008/08/06 05:26:11 org.apache.coyote.http11.Http11Protocol init
INFO: Initializing Coyote HTTP/1.1 on http-8080
2008/08/06 05:26:11 org.apache.catalina.startup.Catalina load
INFO: Initialization processed in 800 ms
2008/08/06 05:26:11 org.apache.catalina.core.StandardService start
INFO: Starting service Catalina
2008/08/06 05:26:11 org.apache.catalina.core.StandardEngine start
INFO: Starting Servlet Engine: Apache Tomcat/6.0.16
2008/08/06 05:26:12 org.apache.catalina.core.StandardContext
addApplicationListener
INFO: The listener "za.co.cocosoft.etude.server.Lifecycle" is already
configured for this context. The duplicate definition has been ignored.
2008/08/06 05:26:12 za.co.cocosoft.etude.server.Lifecycle contextInitialized
CONFIG: Initialising context at /etude
2008/08/06 05:26:12 za.co.cocosoft.etude.scheduler.Scheduler 
INFO: Starting Scheduler Thread...
2008/08/06 05:26:12 za.co.cocosoft.etude.server.Lifecycle initScheduler
CONFIG: Registering Scheduler with ServletContext...
2008/08/06 05:26:13 org.apache.catalina.core.StandardContext
addApplicationListener
INFO: The listener "listeners.ContextListener" is already configured for this
context. The duplicate definition has been ignored.
2008/08/06 05:26:13 org.apache.catalina.core.StandardContext
addApplicationListener
INFO: The listener "listeners.SessionListener" is already configured for this
context. The duplicate definition has been ignored.
2008/08/06 05:26:14 org.apache.coyote.http11.Http11Protocol start
INFO: Starting Coyote HTTP/1.1 on http-8080
2008/08/06 05:26:15 org.apache.catalina.startup.Catalina start
INFO: Server startup in 3243 ms
2008/08/06 05:26:16 za.co.cocosoft.etude.server.Lifecycle contextDestroyed
CONFIG: Destroying context at /etude
2008/08/06 05:26:16 za.co.cocosoft.etude.scheduler.Scheduler terminate
INFO: Shutting Down Scheduler...
2008/08/06 05:26:16 za.co.cocosoft.etude.scheduler.Scheduler terminate
FINE: The Scheduler has left the building!


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 44423] invalid duplicate listeners warning

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=44423


Mark Thomas <[EMAIL PROTECTED]> changed:

   What|Removed |Added

 Status|REOPENED|RESOLVED
 Resolution||FIXED




--- Comment #6 from Mark Thomas <[EMAIL PROTECTED]>  2008-08-06 08:59:56 PST ---
You are not seeing the bug. You are seeing the results of a configuration issue
with your Tomcat instance and/or your application. This is confirmed by the
presence of the context destroy which is only called when the context is being
shut down.

Please use the users list to debug your configuration.


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 45580] Unhandled NullPointerException when trying to write to a closed socket.

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=45580


Filip Hanik <[EMAIL PROTECTED]> changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||INVALID




--- Comment #1 from Filip Hanik <[EMAIL PROTECTED]>  2008-08-06 13:54:23 PST ---
The NPE is only because of an incorrectly used response object. if you are
trying to write to a response that has been closed, in a different thread, it
is better for you to fix the application that is doing that. 

There are simply too many NPE checks to do to catch all such scenarios, and
they essentially don't do anything, they would just convert the NPE to a
IOException.

If you believe the error comes from correct usage of the response, then please
elaborate a little bit on how this happens, and we can apply a correct fix.

best
Filip


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



catalina.sh bug when JULI is not used

2008-08-06 Thread Ian Ward Comfort
I believe commit 652629 (which appeared in 6.0.17) broke catalina.sh  
startup when JULI is not used -- that is, when $CATALINA_BASE/conf/ 
logging.properties does not exist.  If this file is absent,  
LOGGING_CONFIG is not set.  However the variable appears quoted in  
the shell command to exec the JVM:


"$_RUNJAVA" $JAVA_OPTS "$LOGGING_CONFIG" $CATALINA_OPTS ...

Which introduces an empty argument when LOGGING_CONFIG is unset, and  
the JVM interprets this empty argument as the class to be invoked.


Three possible fixes are:

* Set LOGGING_CONFIG to a harmless flag if JULI is not used.  With my  
Sun JVM, a plain -D is safe.  I don't know how portable this is.  It  
isn't very pretty.


* Maintain separate JVM invocations for the cases when LOGGING_CONFIG  
is set and when it's unset.  This involves more code duplication.   
The code in question is already duplicated; I'm not sure whether that  
argues for or against this strategy.


* Use a one-item shell array with an @ subscript, which is immune to  
word splitting when between double quotes but expands to nothing when  
unpopulated.  For example:


	LOGGING_CONFIG[0]="-Djava.util.logging.config.file=$CATALINA_BASE/ 
conf/logging.properties"


followed by:

"$_RUNJAVA" $JAVA_OPTS "[EMAIL PROTECTED]" $CATALINA_OPTS ...

This method may be too clever.  I think POSIX requires shell arrays  
but being portable to true Bourne shell would require $@, which is  
already being used.


I'd be willing to write up a patch if any of these methods seem  
acceptable.


--
Ian Ward Comfort <[EMAIL PROTECTED]>
System Administrator, Student Computing, Stanford University


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



DO NOT REPLY [Bug 45585] New: Tomcat doesn't start when JULI is not used (NoClassDefFoundError)

2008-08-06 Thread bugzilla
https://issues.apache.org/bugzilla/show_bug.cgi?id=45585

   Summary: Tomcat doesn't start when JULI is not used
(NoClassDefFoundError)
   Product: Tomcat 6
   Version: 6.0.18
  Platform: All
OS/Version: All
Status: NEW
  Severity: normal
  Priority: P2
 Component: Catalina
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


If there is no file CATALINA_BASE/conf/logging.properties, then Tomcat will not
start and the JVM outputs the following error:

Exception in thread "main" java.lang.NoClassDefFoundError:

without any additional information.

Quoting from a user's mail:

I believe commit 652629 (which appeared in 6.0.17) broke catalina.sh startup
when JULI is not used -- that is, when $CATALINA_BASE/conf/logging.properties
does not exist.  If this file is absent, LOGGING_CONFIG is not set.  However
the variable appears quoted in the shell command to exec the JVM:

"$_RUNJAVA" $JAVA_OPTS "$LOGGING_CONFIG" $CATALINA_OPTS ...

Which introduces an empty argument when LOGGING_CONFIG is unset, and the JVM
interprets this empty argument as the class to be invoked.

Three possible fixes are:

* Set LOGGING_CONFIG to a harmless flag if JULI is not used.  With my Sun JVM,
a plain -D is safe.  I don't know how portable this is.  It isn't very pretty.

* Maintain separate JVM invocations for the cases when LOGGING_CONFIG is set
and when it's unset.  This involves more code duplication.  The code in
question is already duplicated; I'm not sure whether that argues for or against
this strategy.

* Use a one-item shell array with an @ subscript, which is immune to word
splitting when between double quotes but expands to nothing when unpopulated. 
For example:

   
LOGGING_CONFIG[0]="-Djava.util.logging.config.file=$CATALINA_BASE/conf/logging.properties"

followed by:

"$_RUNJAVA" $JAVA_OPTS "[EMAIL PROTECTED]" $CATALINA_OPTS ...

This method may be too clever.  I think POSIX requires shell arrays but being
portable to true Bourne shell would require $@, which is already being used.


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: catalina.sh bug when JULI is not used

2008-08-06 Thread Rainer Jung

Thank you for reporting the problem. I added you message to Bugilla:

https://issues.apache.org/bugzilla/show_bug.cgi?id=45585

We were already thinking about your first two solutions, I missed the 
third one. We'll need to check how reliable that is (w.r.t. POSIX). I am 
a little in favor of "-Dnop" or something similar.


Regards,

Rainer

Ian Ward Comfort wrote:
I believe commit 652629 (which appeared in 6.0.17) broke catalina.sh 
startup when JULI is not used -- that is, when 
$CATALINA_BASE/conf/logging.properties does not exist.  If this file is 
absent, LOGGING_CONFIG is not set.  However the variable appears quoted 
in the shell command to exec the JVM:


"$_RUNJAVA" $JAVA_OPTS "$LOGGING_CONFIG" $CATALINA_OPTS ...

Which introduces an empty argument when LOGGING_CONFIG is unset, and the 
JVM interprets this empty argument as the class to be invoked.


Three possible fixes are:

* Set LOGGING_CONFIG to a harmless flag if JULI is not used.  With my 
Sun JVM, a plain -D is safe.  I don't know how portable this is.  It 
isn't very pretty.


* Maintain separate JVM invocations for the cases when LOGGING_CONFIG is 
set and when it's unset.  This involves more code duplication.  The code 
in question is already duplicated; I'm not sure whether that argues for 
or against this strategy.


* Use a one-item shell array with an @ subscript, which is immune to 
word splitting when between double quotes but expands to nothing when 
unpopulated.  For example:


LOGGING_CONFIG[0]="-Djava.util.logging.config.file=$CATALINA_BASE/conf/logging.properties" 



followed by:

"$_RUNJAVA" $JAVA_OPTS "[EMAIL PROTECTED]" $CATALINA_OPTS ...

This method may be too clever.  I think POSIX requires shell arrays but 
being portable to true Bourne shell would require $@, which is already 
being used.


I'd be willing to write up a patch if any of these methods seem acceptable.

--
Ian Ward Comfort <[EMAIL PROTECTED]>
System Administrator, Student Computing, Stanford University


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]