[jira] Reopened: (JAMES-580) NPE is issued when receiving a read receipt from MS Outlook, and checkValidSenderDomain is set to true

2006-07-28 Thread Vincenzo Gianferrari Pini (JIRA)
 [ http://issues.apache.org/jira/browse/JAMES-580?page=all ]

Vincenzo Gianferrari Pini reopened JAMES-580:
-

  Assignee: Vincenzo Gianferrari Pini  (was: Norman Maurer)
 
The fix was ok for 3.0 in trunk, but not for 2.3.0.

 NPE is issued when receiving a read receipt from MS Outlook, and 
 checkValidSenderDomain is set to true
 --

 Key: JAMES-580
 URL: http://issues.apache.org/jira/browse/JAMES-580
 Project: James
  Issue Type: Bug
  Components: SMTPServer
Affects Versions: 2.3.0rc1
Reporter: Vincenzo Gianferrari Pini
 Assigned To: Vincenzo Gianferrari Pini
Priority: Blocker
 Fix For: 2.3.0rc2


 A NPE is issued when receiving a read receipt from MS Outlook (not Outlook 
 express nor Thunderbird), and checkValidSenderDomain is set to true and  
 the sender IP address is not in authorizedAddresses:
 27/07/06 17:17:00 ERROR smtpserver: Exception opening socket: null
 java.lang.NullPointerException
   at 
 org.apache.james.smtpserver.MailCmdHandler.doMAIL(MailCmdHandler.java:210)
   at 
 org.apache.james.smtpserver.MailCmdHandler.onCommand(MailCmdHandler.java:83)
   at 
 org.apache.james.smtpserver.SMTPHandler.handleConnection(SMTPHandler.java:391)
   at 
 org.apache.james.util.connection.ServerConnection$ClientConnectionRunner.run(ServerConnection.java:422)
   at 
 org.apache.excalibur.thread.impl.ExecutableRunnable.execute(ExecutableRunnable.java:55)
   at 
 org.apache.excalibur.thread.impl.WorkerThread.run(WorkerThread.java:116)

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira



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



svn commit: r426542 - /james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java

2006-07-28 Thread vincenzo
Author: vincenzo
Date: Fri Jul 28 07:09:31 2006
New Revision: 426542

URL: http://svn.apache.org/viewvc?rev=426542view=rev
Log:
The previous fix to JAMES-580 was ok for 3.0 in trunk, but not for 2.3.0rc1.

Modified:

james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java

Modified: 
james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java
URL: 
http://svn.apache.org/viewvc/james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java?rev=426542r1=426541r2=426542view=diff
==
--- 
james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java
 (original)
+++ 
james/server/branches/v2.3/src/java/org/apache/james/smtpserver/MailCmdHandler.java
 Fri Jul 28 07:09:31 2006
@@ -197,8 +197,8 @@
 }
 }
 
-// check only if its no null sender
-if (checkValidSenderDomain == true  sender != null) {
+// check only if senderAddress is not null
+if (checkValidSenderDomain == true  senderAddress != null) {
 
 /**
  * don't check if the ip address is allowed to relay. Only 
check if it is set in the config. 



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



Re: [jira] Reopened: (JAMES-580) NPE is issued when receiving a read receipt from MS Outlook, and checkValidSenderDomain is set to true

2006-07-28 Thread Norman Maurer
Ups.. :-( Thx
Am Freitag, den 28.07.2006, 07:06 -0700 schrieb Vincenzo Gianferrari
Pini (JIRA):
  [ http://issues.apache.org/jira/browse/JAMES-580?page=all ]
 
 Vincenzo Gianferrari Pini reopened JAMES-580:
 -
 
   Assignee: Vincenzo Gianferrari Pini  (was: Norman Maurer)
  
 The fix was ok for 3.0 in trunk, but not for 2.3.0.
 
  NPE is issued when receiving a read receipt from MS Outlook, and 
  checkValidSenderDomain is set to true
  --
 
  Key: JAMES-580
  URL: http://issues.apache.org/jira/browse/JAMES-580
  Project: James
   Issue Type: Bug
   Components: SMTPServer
 Affects Versions: 2.3.0rc1
 Reporter: Vincenzo Gianferrari Pini
  Assigned To: Vincenzo Gianferrari Pini
 Priority: Blocker
  Fix For: 2.3.0rc2
 
 
  A NPE is issued when receiving a read receipt from MS Outlook (not 
  Outlook express nor Thunderbird), and checkValidSenderDomain is set to 
  true and  the sender IP address is not in authorizedAddresses:
  27/07/06 17:17:00 ERROR smtpserver: Exception opening socket: null
  java.lang.NullPointerException
  at 
  org.apache.james.smtpserver.MailCmdHandler.doMAIL(MailCmdHandler.java:210)
  at 
  org.apache.james.smtpserver.MailCmdHandler.onCommand(MailCmdHandler.java:83)
  at 
  org.apache.james.smtpserver.SMTPHandler.handleConnection(SMTPHandler.java:391)
  at 
  org.apache.james.util.connection.ServerConnection$ClientConnectionRunner.run(ServerConnection.java:422)
  at 
  org.apache.excalibur.thread.impl.ExecutableRunnable.execute(ExecutableRunnable.java:55)
  at 
  org.apache.excalibur.thread.impl.WorkerThread.run(WorkerThread.java:116)
 


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil


[jira] Resolved: (JAMES-580) NPE is issued when receiving a read receipt from MS Outlook, and checkValidSenderDomain is set to true

2006-07-28 Thread Vincenzo Gianferrari Pini (JIRA)
 [ http://issues.apache.org/jira/browse/JAMES-580?page=all ]

Vincenzo Gianferrari Pini resolved JAMES-580.
-

Resolution: Fixed

The fix now is checking senderAddress != null, instead of sender != null.

 NPE is issued when receiving a read receipt from MS Outlook, and 
 checkValidSenderDomain is set to true
 --

 Key: JAMES-580
 URL: http://issues.apache.org/jira/browse/JAMES-580
 Project: James
  Issue Type: Bug
  Components: SMTPServer
Affects Versions: 2.3.0rc1
Reporter: Vincenzo Gianferrari Pini
 Assigned To: Vincenzo Gianferrari Pini
Priority: Blocker
 Fix For: 2.3.0rc2


 A NPE is issued when receiving a read receipt from MS Outlook (not Outlook 
 express nor Thunderbird), and checkValidSenderDomain is set to true and  
 the sender IP address is not in authorizedAddresses:
 27/07/06 17:17:00 ERROR smtpserver: Exception opening socket: null
 java.lang.NullPointerException
   at 
 org.apache.james.smtpserver.MailCmdHandler.doMAIL(MailCmdHandler.java:210)
   at 
 org.apache.james.smtpserver.MailCmdHandler.onCommand(MailCmdHandler.java:83)
   at 
 org.apache.james.smtpserver.SMTPHandler.handleConnection(SMTPHandler.java:391)
   at 
 org.apache.james.util.connection.ServerConnection$ClientConnectionRunner.run(ServerConnection.java:422)
   at 
 org.apache.excalibur.thread.impl.ExecutableRunnable.execute(ExecutableRunnable.java:55)
   at 
 org.apache.excalibur.thread.impl.WorkerThread.run(WorkerThread.java:116)

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira



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



Fix or enhancement to checkResolvableHelo/Ehlo?

2006-07-28 Thread Vincenzo Gianferrari Pini
I'm experimenting (using 2.3.0rc1 in my production system) some new 
settings, like checkResolvableHelo and checkResolvableEhlo.


I found them very effective in blocking spam during the smtp session, in 
addition to blacklists. But I had to drop them because they check also 
the helo provided by SMTP AUTHenticated client users, that may typically 
have non resolvable names specially if roaming.


I think that this check should be conditioned by the user being or not 
SMTP AUTHenticated, as the blacklist check is in RcptCmdHandler, so IMHO 
the current behaviour is a bug that should be fixed for 2.3.0. But as 
this whole helo/ehlo check functionality was introduced some time ago by 
Stefano, he may have given it a different meaning then mine, so I'm not 
sure of it.


What do you think about this?

Vincenzo



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



Re: Fix or enhancement to checkResolvableHelo/Ehlo?

2006-07-28 Thread Norman Maurer
Am Freitag, den 28.07.2006, 17:22 +0200 schrieb Vincenzo Gianferrari
Pini:
 I'm experimenting (using 2.3.0rc1 in my production system) some new 
 settings, like checkResolvableHelo and checkResolvableEhlo.
 
 I found them very effective in blocking spam during the smtp session, in 
 addition to blacklists. But I had to drop them because they check also 
 the helo provided by SMTP AUTHenticated client users, that may typically 
 have non resolvable names specially if roaming.
 
 I think that this check should be conditioned by the user being or not 
 SMTP AUTHenticated, as the blacklist check is in RcptCmdHandler, so IMHO 
 the current behaviour is a bug that should be fixed for 2.3.0. But as 
 this whole helo/ehlo check functionality was introduced some time ago by 
 Stefano, he may have given it a different meaning then mine, so I'm not 
 sure of it.
 
 What do you think about this?
 
 Vincenzo

Well Vincenzo you are right. i will correct it. 
BTW the code was contributed from me and stefano just commit it.. So its
my faul :-(

bye
Norman 


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil


Re: Fix or enhancement to checkResolvableHelo/Ehlo?

2006-07-28 Thread Stefano Bagnara

Vincenzo Gianferrari Pini wrote:
I found them very effective in blocking spam during the smtp session, in 
addition to blacklists. But I had to drop them because they check also 
the helo provided by SMTP AUTHenticated client users, that may typically 
have non resolvable names specially if roaming.


They are effective but even dungerous. MANY users have misconfigured 
systems and uses bad salutations.
Furthermore the EHLO/EHLO is done before the AUTH can be issued so we 
only have a checkAuthNetworks option and not a checkAuthClients 
option like we did for other CmdHandlers.


I think that this check should be conditioned by the user being or not 
SMTP AUTHenticated, as the blacklist check is in RcptCmdHandler, so IMHO 
the current behaviour is a bug that should be fixed for 2.3.0. But as 
this whole helo/ehlo check functionality was introduced some time ago by 
Stefano, he may have given it a different meaning then mine, so I'm not 
sure of it.


I don't remember if I committed it or not, btw this is Norman work 
(IIRC). I never worked on this ehlo validity/resolvability checks 
because I think they are more harmful than useful.


I've not checked the sources and I don't know how it works now. I 
suppose you already checked the checkAuthClients/checkAuthNetworks 
option for the handlers.



What do you think about this?

Vincenzo


I am against any change to the fastfail code because we are in RC and 
the fastfail is experimental code that we already changed it in trunk 
and should not delay furthermore 2.3.0 for this stuff. (In fact I would 
have released 2.3.0 with fastfail disabled at all and only document how 
to enable it somewhere in our wiki for people that wanted to experiment).


I think I will not veto this changes, but I think that if we start 
fixing this behaviours we'll never make this release.


Stefano


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



RE: Fix or enhancement to checkResolvableHelo/Ehlo?

2006-07-28 Thread Noel J. Bergman
Stefano Bagnara wrote:

 the EHLO/EHLO is done before the AUTH can be issued so we only
 have a checkAuthNetworks option and not a checkAuthClients
 option like we did for other CmdHandlers.

Sometimes, you have to defer the problem that you caught, so if the command
handler finds the problem, it can set a session attribute itself (or another
aware handler) to deal with later.

For example, the block list code is intentionally handled in RCPT because I
wanted to defer the error until we knew to whom they were sending the
e-mail, thus allowing RFC compliance with postmaster@ and [EMAIL PROTECTED]

The concept, for which we now have support in the v2.4 changes, was that an
EHLO handler could actually be a suite of related handlers registered to
be an EHLO handler, an AUTH handler, and some other type of handler(s), so
that it could set a detected concern in EHLO, possibly clear it in AUTH, and
check it downstream.

I've not formed an opinion on whether or not to fix this for v2.3 or wait
for v2.4.  I won't form an opinion until I see the proposed patch.

--- Noel


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



Postage problems and questions

2006-07-28 Thread Vincenzo Gianferrari Pini

Bernd,

I'm playing with Postage and 2.3.0rc1: Postage is great! Thanks.

I found a couple of problems:

1) Postage doesn't support SMTP Authentication for int-ext users. 
Perhaps it is expected, as a real config.xml has in any case to be 
hacked to run postage, and setting authRequired to false is easy. But 
insuch case it should be said in the docs.


2) The generated postage_jvmStatistics.*.csv (and postage_errors.*.csv) 
files are empty after a run.


Now a question: the time interval between any two mails generated (based 
on count-per-min=n) is constant or is there any probability 
distribution function (exponential or other)? And what about the text 
and binary sizes?


Thanks,

Vincenzo


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



RE: Conditional build / check for jdbc 3.x presence

2006-07-28 Thread Noel J. Bergman
Yes, I believe that we can remove it.  We can also removed some other
things, such as the changes Norman made, as per comments in the code, to
remove workarounds for earlier JDKs.

--- Noel


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



RE: svn commit: r426107 - in /james/server/trunk/src/java/org/apache/james: transport/mailets/CommandListservProcessor.java transport/mailets/GenericListserv.java transport/mailets/WhiteListManager.ja

2006-07-28 Thread Noel J. Bergman
Norman,

My major issue with this commit is that you preserved the comment:

+ * pI have done extensive testing of this routine with a standalone
+ * driver, and am leaving the commented out debug messages so that
+ * when someone decides to enhance this method, it can be yanked it
+ * from this file, embedded it with a test driver, and the comments
+ * enabled./p

But you removed all of the debug code!  One or other other should be corrected: 
either the comment should be removed, or the debug code put back.  :-)

--- Noel


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