Re: cvs commit: jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5CoyoteAdapter.java

2003-07-03 Thread Remy Maucherat
[EMAIL PROTECTED] wrote:
keith   2003/07/02 17:16:49

  Modified:catalina/src/share/org/apache/coyote/tomcat5
CoyoteAdapter.java
  Log:
  Allow * as a valid url
Errr, ok, well, that's a possibly risky patch. I think that does get 
mapped to the root context and its default servlet (and the servlet path 
should be *).
I verified a /* security constraint mapping matched this URL.

Remy

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


Re: cvs commit: jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5 CoyoteAdapter.java

2003-07-03 Thread Bill Barker

- Original Message -
From: Remy Maucherat [EMAIL PROTECTED]
To: Tomcat Developers List [EMAIL PROTECTED]
Sent: Wednesday, July 02, 2003 11:49 PM
Subject: Re: cvs commit:
jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5
CoyoteAdapter.java


 [EMAIL PROTECTED] wrote:
  keith   2003/07/02 17:16:49
 
Modified:catalina/src/share/org/apache/coyote/tomcat5
  CoyoteAdapter.java
Log:
Allow * as a valid url

 Errr, ok, well, that's a possibly risky patch. I think that does get
 mapped to the root context and its default servlet (and the servlet path
 should be *).
 I verified a /* security constraint mapping matched this URL.


I've checked this briefly.  It looks Ok to me.  The original problem was
with:
$ telnet localhost 8080
OPTIONS * /HTTP1.1
Host: localhost:8080



Previously this would generate a 400 error.  Now it generates the correct
values for the OPTIONS with a (correct) 200 response.


 Remy


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



This message is intended only for the use of the person(s) listed above as the 
intended recipient(s), and may contain information that is PRIVILEGED and 
CONFIDENTIAL.  If you are not an intended recipient, you may not read, copy, or 
distribute this message or any attachment. If you received this communication in 
error, please notify us immediately by e-mail and then delete all copies of this 
message and any attachments.

In addition you should be aware that ordinary (unencrypted) e-mail sent through the 
Internet is not secure. Do not send confidential or sensitive information, such as 
social security numbers, account numbers, personal identification numbers and 
passwords, to us via ordinary (unencrypted) e-mail.

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

DO NOT REPLY [Bug 21288] - DirContext caching problem with Dreamweaver 6 when recursively creating directories

2003-07-03 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21288.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21288

DirContext caching problem with Dreamweaver 6 when recursively creating directories





--- Additional Comments From [EMAIL PROTECTED]  2003-07-03 08:57 ---
Remy,
That cache unload would indeed sort the problem. My suggestion was to fix the trailing 
slash issue in WebdavServlet. You could argue that ProxyDirContext is currently doing 
the right thing, treating /some.gif and /some.gif/ as two different entries and that 
is the 
user of the API (WebdavServlet) that should make sure the cache pollution doesn't 
occur. 
But I suppose that's philosophical since the end result would be the same.

I would be happy to test the patch for you. I'm doing TC4.1 on my live systems.
Did you have any thoughts on the #21284 issue?

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



DO NOT REPLY [Bug 21279] - Proc Runner: No Listener Notification on Shutdown

2003-07-03 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21279.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21279

Proc Runner: No Listener Notification on Shutdown





--- Additional Comments From [EMAIL PROTECTED]  2003-07-03 10:12 ---
Some more observation - hope it helps to pin down the problem:
If I start TC via proc runner and stop TC via shutdown.bat the method 
contextDestroyed() is properly called.

Windows Task-Manager:
Stopping TC via proc runner gives a peak CPU utilization of approx. 6%, the 
extra CPU utilization lasts for approx. 5 seconds.
Stopping TC via shutdown.bat gives a peak CPU utilization of 100%, the extra 
CPU utilization lasts from 15 to 35 seconds.

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



how to view structure about Tomcat 4?

2003-07-03 Thread Dan Wang
I use many hours to debug tomcat 4 when it starts.But I can not view whole structure 
about Tomcat 4 .
Where are the documents which describe Tomcat 4 's structure located?
Who can help me ?

Thanks .and thanks

Danwang

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



How to set ServerSocketFactory

2003-07-03 Thread Yuri Schimke
Hi,

I'm trying to set the ServerSocketFactory in Tomcat 4.1.24.  My factory
is getting created (by the catalina digester), but it never actually
gets used by the PoolTcpEndpoint. I can see this from debugging I am
using.  Its using the DefaultServerSocketFactory instead of mine.

0 [main] INFO net.BlahServerSocketFactory  - new
BlahServerSocketFactory()
java.net.BindException: Permission denied
at java.net.PlainSocketImpl.socketBind(Native Method)
at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:331)
at java.net.ServerSocket.bind(ServerSocket.java:318)
at java.net.ServerSocket.init(ServerSocket.java:185)
at net.DefaultServerSocketFactory.createSocket(:102)
at PoolTcpEndpoint.initEndpoint(:277)
at http11.Http11Protocol.init(:150)
at CoyoteConnector.initialize(:1117)


My class is like (I also tried extending
tomcat.util.ServerSocketFactory)

public final class BlahServerSocketFactory 
implements org.apache.catalina.net.ServerSocketFactory {

}

My config is 

Connector className=org.apache.coyote.tomcat4.CoyoteConnector
   address=localhost port=80
   minProcessors=5 maxProcessors=75
   enableLookups=true redirectPort=8443
   acceptCount=100 debug=0 connectionTimeout=6
   useURIValidationHack=false disableUploadTimeout=true
  Factory className=com.zed.tomcat.net.BlahServerSocketFactory /
/Connector

  
I searched for anywhere PoolTcpEndpoint.setServerSocketFactory() gets
called, but I don't think it ever is.

Could someone let me know how I am meant to achieve this?

Thanks.

--
Yuri Schimke
Aqris Software
+372 53 415 579

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



Re: how to view structure about Tomcat 4?

2003-07-03 Thread Yoav Shapira
Howdy,
All the documentation is online at http://jakarta.apache.org/tomcat.  Your
question is more suited to the tomcat-user mailing list, so please ask further
questions along this thread there.

Yoav Shapira

--- Dan Wang [EMAIL PROTECTED] wrote:
 I use many hours to debug tomcat 4 when it starts.But I can not view whole
 structure about Tomcat 4 .
 Where are the documents which describe Tomcat 4 's structure located?
 Who can help me ?
 
 Thanks .and thanks
 
 Danwang
 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 


=
Yoav Shapira
[EMAIL PROTECTED]

__
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



Re: [JAKARTA NEWSLETTER DRAFT] News on 'Jakarta Tomcat' from May to June, 2003

2003-07-03 Thread Yoav Shapira
Howdy,
Good paragraph Remy..

 P.S. Are here anyone who have been voted in as new committers of Tomcat
 within these 2 months?

I've already added myself to your wiki page per your request, and I haven't
checked but I'm sure Tim Funk will get around to it as well ;)

Yoav Shapira


=
Yoav Shapira
[EMAIL PROTECTED]

__
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

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



Logging and Classloading issues

2003-07-03 Thread Martin van Dijken
I've got several problems using commons logging and log4j in Tomcat and I've been 
searching on ways to solve them. The actual problem I have is the following:

org.apache.commons.logging.LogConfigurationException: Class 
org.apache.commons.logging.impl.Log4JCategoryLog does not implement Log

With of course a huge stack trace behind it. Now I found quite a few articles on the 
web and some mails in the tomcat-dev list about this problem but none seem to report a 
fully working solution. What I've setup is the following: 

$Tomcat_home/common/lib 
- no commons-logging nor log4j

$tomcat_home/server/lib 
- left the commons-logging there. Should be fully separate from my webapps

WEB-APP/WEB-INF/lib
- commons-logging 1.0.2
- log4j 1.2.6
- a lib requiring commons-logging

WEB-APP/WEB-INF/classes
- log4j.properties

I've found on this list that Costin solved the above exception by putting 
commons-logging and log4j in the $tomcat_home/common/lib directory and removing them 
from the web-apps entirely. I've tried that setup and it seems to work. The above 
exception is gone. The problem with it however, is that I can't have every web-app use 
different appenders and verbosity settings any more. The only log4j.properties file 
this setup accepts can be on a global location.

So, does anybody know of a solution that allows me to get rid of all of the does not 
implement Log exceptions and allow me to use logging on a per-webapp base?

Martin van Dijken

P.S. It's not a choice of mine to use commons-logging instead of direct log4j. A 
library we use (Kodo)requires this. Therefore I can't get rid of commons-logging as 
Ceki from log4j suggests:
http://www.qos.ch/logging/thinkAgain.html

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



[GUMP] Build timed out - jk

2003-07-03 Thread Craig McClanahan

This email is autogenerated from the output from:
http://cvs.apache.org/builds/gump/2003-07-03/jakarta-tomcat-jk-native.html


Buildfile: build.xml

init:
 [echo] /home/rubys
[mkdir] Created dir: /home/rubys/jakarta/jakarta-tomcat-connectors/jk/build/jk
 [echo] linux=true solaris=${solaris} win32=${win32} hpux=${hpux} 
netware=${netware}

apache20:

apache13:

iis:

netscape:

jni:
[mkdir] Created dir: /home/rubys/jakarta/jakarta-tomcat-connectors/jk/build/jk/jni
   [so] Compiling 4 out of 4
Compiling /home/rubys/jakarta/jakarta-tomcat-connectors/jk/native/common/jk_map.c
Compiling /home/rubys/jakarta/jakarta-tomcat-connectors/jk/native/common/jk_pool.c
/home/rubys/bin/timeout: timed out

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



RE: How to get the container within an instance event?

2003-07-03 Thread Angus Mezick
Cool! That is exactly what I needed!  I was getting miffed because
valves got the nice catalina request object while InstanceEvent had to
deal with the dumbed down HttpRequest.  Now onto a design question:

I am implementing a JdbcSessionManager for dealing with non-sticky
clusters.  We wish to use the DB to store all of our session data when
it is not in use.  Thanks to Craig's help I was able to figure out how
to do the updating of the session data in an InstanceEventListener that
catches the AFTER_SERVICE_EVENT.  In talking to Filip Hank about his
Clustered session manager I was also able to do the database updates in
a valve.  Which way is preferable?  In the InstanceEvent I can just
filter out all queries that make this true:
default.equals(event.getServlet().getServletConfig().getServletName())
.  Filip used the file extensions to create an exclusion list.

Other notes:
I have had to create my own JdbcSession object instead of subclassing
StandardSession because the attributes HashMap is private and I need
access to it.  (I think)  I am serializing the HashMap and storing only
that in its own column.  My other columns are expire_time,
creation_time, app_name, and session_id.  I am currently limited to
sessions with 8k of data because of the DB limits on VARBINARY but that
can change at a later date.

--Angus

 -Original Message-
 From: Craig R. McClanahan [mailto:[EMAIL PROTECTED] 
 Sent: Wednesday, July 02, 2003 11:50 PM
 On Wed, 2 Jul 2003, Angus Mezick wrote:
 
  Date: Wed, 2 Jul 2003 15:08:33 -0400
  From: Angus Mezick [EMAIL PROTECTED]
  Does anyone know how to access the container object from an instance
  event?  Are there any specs that say an instance event can have no
  internal access to the server?
 
 
 I presume you're talking about 
 org.apache.catalina.InstanceEvent, right?
 If so, the wrapper property that is accessible via 
 getWrapper() is the
 org.apache.catalina.Container instance for this particular 
 servlet.  If
 you want the Context element for the webapp, for example, you'd say
 something like:
 
   Context context = (Context) event.getWrapper().getParent();
 
 or, to get the Host for this virtual host:
 
   Host host = (Host) event.getWrapper().getParent().getParent();
 
 and so on.
 
  Sorry for the cross post but I am deep into the internals 
 of tomcat here
  and I don't know which list is appropriate.  The list with how do I
  change the default port type q's or the one that debates 
 how features
  should be implemented.
 
 
 The TOMCAT-USER list probably won't have many folks that 
 understand the
 internals well enough to answer a question like this :-).
 
  Angus Mezick
  GuideStar - Philanthropic Research Inc.
 Craig McClanahan

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



WebdavServlet.java, form based login and files from other contexts

2003-07-03 Thread Martin Algesten
In a recent client project I used the WebdavServlet for the first time 
and encountered some configuration issues which lead me down the path 
of doing some patches to it.

Our client's site uses form based login which doesn't play well with 
WebDAV. All clients I've ever seen only do basic authentication. The 
problem is really that one web app can only have one type of login (I 
would like to be able to set different login types for different 
security-constraints in the same app). Changing to basic login is not 
really an option since that makes the site less slick, and part of the 
project was explicitly to do form based login.

The only way I can see to solve the problem was to add another 
parameter into the WebdavServlet so that it can server files of another 
context, in my configuration a /webdav webapp serves files off the 
ROOT (/) context.

The changes to the WebdavServlet to do this are very trivial, add 
another parameter fileContext and override the getResources() method 
in the DefaultServlet with an exact copy except that rather than 
grabbing the local context, the method can now grab a foreign one if 
fileContext is set.

The added bonus with this solution is that .jsp files can be served 
from the the ROOT context through jasper, while the source code can be 
served in my /webdav context.

Is this an okay solution? Is it the right way of doing it? ... 
Shouldn't really the WebdavServlet be able to serve an arbitrary path 
of the file system? Have anyone else had the same problem, should the 
WebdavServlet be possible to configure along these lines?

Martin

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


WebdavServlet and security constraints

2003-07-03 Thread Martin Algesten
In a recent project we cooperated with our client in building a site. 
They were doing static html files and we took some of them and turned 
into JSPs.

In order for our client to be able to update their site, we enabled 
webdav on the webapplication. However, we don't want them to be able to 
update our JSPs which lives in the same webapp as their static content. 
We see it as a security risk to be able to write arbitrary JSP code 
that gets executed on our servers.

The neatest way of doing this protection was in the web.xml:

!-- This constraint makes sure no JSP files are overwritten but they 
can
 still be downloaded. Only dangerous http-methods are forbidden. 
--
  security-constraint
web-resource-collection
  web-resource-nameForbidden/web-resource-name
  url-pattern*.jsp/url-pattern
!--  http-methodGET/http-method --
  http-methodPOST/http-method
  http-methodPUT/http-method
  http-methodDELETE/http-method
!--  http-methodOPTIONS/http-method --
  http-methodPATCH/http-method
!--  http-methodPROPFIND/http-method --
  http-methodPROPPATCH/http-method
  http-methodMKCOL/http-method
  http-methodCOPY/http-method
  http-methodMOVE/http-method
  http-methodLOCK/http-method
  http-methodUNLOCK/http-method
/web-resource-collection
auth-constraint
  !-- No roles, which means no one is allowed --
/auth-constraint

Which works a charm, however it does not protect us for MOVE 
operations. The webdav client can still create a file such as file.txt 
and rename it to file.jsp and then execute it. This is because when 
renaming file.txt to something else the HTTP request issues is MOVE 
/file.txt and only later in the XML blurb is the destination file 
specified.

Now really what I would like to do is to make a move filea.txt 
fileb.jsp from a security point to be equal to a DELETE filea.txt 
and PUT fileb.jsp. If the user is allowed to do these two operations, 
then he/she is allowed to do the MOVE.

Is there a nice server side way of doing this security check? I don't 
think the servlet API exposes anything useful. The only generic way of 
doing it that I can think of is constructing URL objects that calls my 
own servlet (and of course the added grief of copying over security 
headers etc) with some URL parameter that tells the servlet that the 
operation is only for test.

The other way is to do a TC specific solution, if I only could get hold 
of the internal APIs I could easily check this. I found the 
SecurityConstraint object which resides in the StandardContext object. 
However these classes seems very internal to TC and I suspect there is 
no good way of getting hold of them from the WebdavServlet, right?

Martin

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


RE: class diagram

2003-07-03 Thread Shapira, Yoav

Howdy,
Download tomcat source and use one of these tools:
http://www.cse.ucsc.edu/classes/cmps115/Fall02/tools/tools.html

Yoav Shapira
Millennium ChemInformatics


-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
Sent: Thursday, July 03, 2003 1:24 AM
To: Tomcat Developers List
Subject: class diagram


I was asking abt UML class diagram, not API's.
Could u please tell me where can i get it.

- Kaushik








http://jakarta.apache.org/tomcat/tomcat-4.1-
doc/catalina/docs/api/index.html



 Hi All,


 From where can I get complete class diagrams of Tomcat ?



 Thanks,

 Kaushik











This e-mail, including any attachments, is a confidential business communication, and 
may contain information that is confidential, proprietary and/or privileged.  This 
e-mail is intended only for the individual(s) to whom it is addressed, and may not be 
saved, copied, printed, disclosed or used by anyone else.  If you are not the(an) 
intended recipient, please immediately delete this e-mail from your computer system 
and notify the sender.  Thank you.


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



[PATCH] CATALINA_LOGDIR environment variable in catalina.sh

2003-07-03 Thread Thorsten Huber
Our Tomcat setup requires that the log files Tomcat generates reside
outside CATALINA_BASE and CATALINA_HOME. So I've defined a new
environment variable in catalina.sh named CATALINA_LOGDIR. With this
it is easier to set up different Tomcat installations with different
log directories.

If I'm not the only one who has this problem, it would be great if
this change gets included into the next Tomcat release, so we don't
have to patch our Tomcat installations on future upgrades.

I've attached a patch against Tomcat 4.1.24

best regards
Thorsten Huber

--- catalina.sh.origThu Jul  3 13:34:01 2003
+++ catalina.sh Thu Jul  3 14:36:28 2003
@@ -17,6 +17,9 @@
 #   the JVM should use (java.io.tmpdir).  Defaults to
 #   $CATALINA_BASE/temp.
 #
+#   CATALINA_LOGDIR (Optional) Directory path location of log directory for
+#   catalina.out
+#
 #   JAVA_HOME   Must point at your Java Development Kit installation.
 #
 #   JAVA_OPTS   (Optional) Java runtime options used when the start,
@@ -98,6 +101,11 @@
   CATALINA_TMPDIR=$CATALINA_BASE/temp
 fi
 
+if [ -z CATALINA_LOGDIR ] ; then
+  # Define the default log directory
+  CATALINA_LOGDIR=$CATALINA_BASE/logs
+fi
+
 # For Cygwin, switch paths to Windows format before running java
 if $cygwin; then
   JAVA_HOME=`cygpath --path --windows $JAVA_HOME`
@@ -113,6 +121,7 @@
 echo Using CATALINA_BASE:   $CATALINA_BASE
 echo Using CATALINA_HOME:   $CATALINA_HOME
 echo Using CATALINA_TMPDIR: $CATALINA_TMPDIR
+echo Using CATALINA_LOGDIR: $CATALINA_LOGDIR
 echo Using JAVA_HOME:   $JAVA_HOME
 
 if [ $1 = jpda ] ; then
@@ -191,7 +200,7 @@
 elif [ $1 = start ] ; then
 
   shift
-  touch $CATALINA_BASE/logs/catalina.out
+  touch $CATALINA_LOGDIR/catalina.out
   if [ $1 = -security ] ; then
 echo Using Security Manager
 shift
@@ -203,7 +212,7 @@
   -Dcatalina.home=$CATALINA_HOME \
   -Djava.io.tmpdir=$CATALINA_TMPDIR \
   org.apache.catalina.startup.Bootstrap $@ start \
-   $CATALINA_BASE/logs/catalina.out 21 
+   $CATALINA_LOGDIR/catalina.out 21 
 
   if [ ! -z $CATALINA_PID ]; then
 echo $!  $CATALINA_PID
@@ -215,7 +224,7 @@
   -Dcatalina.home=$CATALINA_HOME \
   -Djava.io.tmpdir=$CATALINA_TMPDIR \
   org.apache.catalina.startup.Bootstrap $@ start \
-   $CATALINA_BASE/logs/catalina.out 21 
+   $CATALINA_LOGDIR/catalina.out 21 
 
   if [ ! -z $CATALINA_PID ]; then
 echo $!  $CATALINA_PID

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



url doesn't work

2003-07-03 Thread RSEQUEIRA

The following url doesn't work on the tomcat site:
http://jakarta.apache.org/tomcat/tomcat-5.0-doc/jk2/index.html


Roshan Sequeira
314-373-6511


This transmission is intended to be strictly confidential. If you are not
the intended recipient of this message, you may not disclose, print, copy
or disseminate this information. If you have received this in error, please
reply and notify the sender (only) and delete the message. Unauthorized
interception of this e-mail is a violation of federal criminal law.



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



RE: [PATCH] CATALINA_LOGDIR environment variable in catalina.sh

2003-07-03 Thread Shapira, Yoav

Howdy,
That's why the Logger elements supports anything you want in its
directory attribute.  I don't think this patch is strictly necessary
and I'd prefer to keep the startup scripts as simple as possible.  What
does this patch give you that you can't already do?

Yoav Shapira
Millennium ChemInformatics


-Original Message-
From: Thorsten Huber [mailto:[EMAIL PROTECTED]
Sent: Thursday, July 03, 2003 9:47 AM
To: [EMAIL PROTECTED]
Subject: [PATCH] CATALINA_LOGDIR environment variable in catalina.sh

Our Tomcat setup requires that the log files Tomcat generates reside
outside CATALINA_BASE and CATALINA_HOME. So I've defined a new
environment variable in catalina.sh named CATALINA_LOGDIR. With this
it is easier to set up different Tomcat installations with different
log directories.

If I'm not the only one who has this problem, it would be great if
this change gets included into the next Tomcat release, so we don't
have to patch our Tomcat installations on future upgrades.

I've attached a patch against Tomcat 4.1.24

best regards
Thorsten Huber

--- catalina.sh.orig   Thu Jul  3 13:34:01 2003
+++ catalina.shThu Jul  3 14:36:28 2003
@@ -17,6 +17,9 @@
 #   the JVM should use (java.io.tmpdir).  Defaults to
 #   $CATALINA_BASE/temp.
 #
+#   CATALINA_LOGDIR (Optional) Directory path location of log
directory
for
+#   catalina.out
+#
 #   JAVA_HOME   Must point at your Java Development Kit
installation.
 #
 #   JAVA_OPTS   (Optional) Java runtime options used when the
start,
@@ -98,6 +101,11 @@
   CATALINA_TMPDIR=$CATALINA_BASE/temp
 fi

+if [ -z CATALINA_LOGDIR ] ; then
+  # Define the default log directory
+  CATALINA_LOGDIR=$CATALINA_BASE/logs
+fi
+
 # For Cygwin, switch paths to Windows format before running java
 if $cygwin; then
   JAVA_HOME=`cygpath --path --windows $JAVA_HOME`
@@ -113,6 +121,7 @@
 echo Using CATALINA_BASE:   $CATALINA_BASE
 echo Using CATALINA_HOME:   $CATALINA_HOME
 echo Using CATALINA_TMPDIR: $CATALINA_TMPDIR
+echo Using CATALINA_LOGDIR: $CATALINA_LOGDIR
 echo Using JAVA_HOME:   $JAVA_HOME

 if [ $1 = jpda ] ; then
@@ -191,7 +200,7 @@
 elif [ $1 = start ] ; then

   shift
-  touch $CATALINA_BASE/logs/catalina.out
+  touch $CATALINA_LOGDIR/catalina.out
   if [ $1 = -security ] ; then
 echo Using Security Manager
 shift
@@ -203,7 +212,7 @@
   -Dcatalina.home=$CATALINA_HOME \
   -Djava.io.tmpdir=$CATALINA_TMPDIR \
   org.apache.catalina.startup.Bootstrap $@ start \
-   $CATALINA_BASE/logs/catalina.out 21 
+   $CATALINA_LOGDIR/catalina.out 21 

   if [ ! -z $CATALINA_PID ]; then
 echo $!  $CATALINA_PID
@@ -215,7 +224,7 @@
   -Dcatalina.home=$CATALINA_HOME \
   -Djava.io.tmpdir=$CATALINA_TMPDIR \
   org.apache.catalina.startup.Bootstrap $@ start \
-   $CATALINA_BASE/logs/catalina.out 21 
+   $CATALINA_LOGDIR/catalina.out 21 

   if [ ! -z $CATALINA_PID ]; then
 echo $!  $CATALINA_PID

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




This e-mail, including any attachments, is a confidential business communication, and 
may contain information that is confidential, proprietary and/or privileged.  This 
e-mail is intended only for the individual(s) to whom it is addressed, and may not be 
saved, copied, printed, disclosed or used by anyone else.  If you are not the(an) 
intended recipient, please immediately delete this e-mail from your computer system 
and notify the sender.  Thank you.


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



RE: url doesn't work

2003-07-03 Thread Shapira, Yoav

Howdy,
Thanks - there's already an open Bugzilla item for this and it'll get
fixed ;)

Yoav Shapira
Millennium ChemInformatics


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Thursday, July 03, 2003 9:39 AM
To: [EMAIL PROTECTED]
Subject: url doesn't work


The following url doesn't work on the tomcat site:
http://jakarta.apache.org/tomcat/tomcat-5.0-doc/jk2/index.html


Roshan Sequeira
314-373-6511


This transmission is intended to be strictly confidential. If you are
not
the intended recipient of this message, you may not disclose, print,
copy
or disseminate this information. If you have received this in error,
please
reply and notify the sender (only) and delete the message. Unauthorized
interception of this e-mail is a violation of federal criminal law.



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




This e-mail, including any attachments, is a confidential business communication, and 
may contain information that is confidential, proprietary and/or privileged.  This 
e-mail is intended only for the individual(s) to whom it is addressed, and may not be 
saved, copied, printed, disclosed or used by anyone else.  If you are not the(an) 
intended recipient, please immediately delete this e-mail from your computer system 
and notify the sender.  Thank you.


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



Re: [PATCH] CATALINA_LOGDIR environment variable in catalina.sh

2003-07-03 Thread Thorsten Huber
On Thu, Jul 03, 2003 at 09:58:32AM -0400, Shapira, Yoav wrote:
 Howdy,
 That's why the Logger elements supports anything you want in its
 directory attribute.  I don't think this patch is strictly necessary
 and I'd prefer to keep the startup scripts as simple as possible.  What
 does this patch give you that you can't already do?

This patch is a way to move the catalina.out to another location than
logs/catalina.out without modifying the catalina.sh on every
installation. If I'm the only one who needs this, then I have to add
these changes on every new tomcat release. But there may be other
people that want to locate the catalina.out in another directory, so
it would be easier to make this location configurable.
 
best regards
Thorsten Huber

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



cvs commit: jakarta-tomcat-connectors/jk/tools/reports tomcat_trend.pl

2003-07-03 Thread glenn
glenn   2003/07/03 07:41:12

  Modified:jk/tools/reports tomcat_trend.pl
  Log:
  Prevent script from failing on bad date format
  
  Revision  ChangesPath
  1.5   +7 -3  jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl
  
  Index: tomcat_trend.pl
  ===
  RCS file: /home/cvs/jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl,v
  retrieving revision 1.4
  retrieving revision 1.5
  diff -u -r1.4 -r1.5
  --- tomcat_trend.pl   14 Mar 2003 13:36:58 -  1.4
  +++ tomcat_trend.pl   3 Jul 2003 14:41:12 -   1.5
  @@ -179,6 +179,10 @@
 ($mon, $day, $time, $year) = (split /\s+/,$line)[1..4];
 ($hour, $min, $sec) = split /:/,$time;
 $year =~ s/\]$//;
  +  if( $day !~ /^\d+/ || $hour !~ /^\d+/ || $min!~ /^\d+/ || $sec !~ /^\d+/ ) {
  + print Unknown log entry: $origline\n unless $origline =~ /\.c /;
  + next;
  +  }
 # print $mon $day $time $year $hour $min $sec\n;
 $logtime = timelocal($sec,$min,$hour,$day,$MON{$mon},$year-1900);
   
  
  
  

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



Re: [PATCH] CATALINA_LOGDIR environment variable in catalina.sh

2003-07-03 Thread Tim Funk
All of the loggers may have an attibute called directory. This may be 
relative or absolute.

No extra variables needed for other scripts - everything can stay confined to 
server.xml.

-Tim

Thorsten Huber wrote:
On Thu, Jul 03, 2003 at 09:58:32AM -0400, Shapira, Yoav wrote:
This patch is a way to move the catalina.out to another location than
logs/catalina.out without modifying the catalina.sh on every
installation. If I'm the only one who needs this, then I have to add
these changes on every new tomcat release. But there may be other
people that want to locate the catalina.out in another directory, so
it would be easier to make this location configurable.
 
best regards
Thorsten Huber
 


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


DO NOT REPLY [Bug 21314] New: - Admin app edit overwrites roles in tomcat-users

2003-07-03 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21314.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=21314

Admin app edit overwrites roles in tomcat-users

   Summary: Admin app edit overwrites roles in tomcat-users
   Product: Tomcat 4
   Version: 4.1.24
  Platform: PC
OS/Version: Windows XP
Status: NEW
  Severity: Normal
  Priority: Other
 Component: Webapps:Administration
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


Changing the Full Name property of a user via the Admin webapp causes the
tomcat-users file to have the roles property removed/overwritten by the full
name property.

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



Re: cvs commit: jakarta-tomcat-connectors/jk/tools/reports tomcat_trend.pl

2003-07-03 Thread Henri Gomez
[EMAIL PROTECTED] a écrit :
glenn   2003/07/03 07:41:12

  Modified:jk/tools/reports tomcat_trend.pl
  Log:
  Prevent script from failing on bad date format
  
  Revision  ChangesPath
  1.5   +7 -3  jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl
  
  Index: tomcat_trend.pl
  ===
  RCS file: /home/cvs/jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl,v
  retrieving revision 1.4
  retrieving revision 1.5
  diff -u -r1.4 -r1.5
  --- tomcat_trend.pl	14 Mar 2003 13:36:58 -	1.4
  +++ tomcat_trend.pl	3 Jul 2003 14:41:12 -	1.5
  @@ -179,6 +179,10 @@
 ($mon, $day, $time, $year) = (split /\s+/,$line)[1..4];
 ($hour, $min, $sec) = split /:/,$time;
 $year =~ s/\]$//;
  +  if( $day !~ /^\d+/ || $hour !~ /^\d+/ || $min!~ /^\d+/ || $sec !~ /^\d+/ ) {
  + print Unknown log entry: $origline\n unless $origline =~ /\.c /;
  + next;
  +  }
 # print $mon $day $time $year $hour $min $sec\n;
 $logtime = timelocal($sec,$min,$hour,$day,$MON{$mon},$year-1900);
   


Will it be included in JK 1.2.5 release ?

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


RE: cvs commit: jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5 CoyoteAdapter.java

2003-07-03 Thread Keith Wannamaker
Hi Remy, * does map to the default context.  Can you think of any special 
downstream handling that we should do for '*'?  

Keith

| -Original Message-
| From: Remy Maucherat [mailto:[EMAIL PROTECTED]
| Sent: Thursday, July 03, 2003 2:50 AM
| To: Tomcat Developers List
| Subject: Re: cvs commit:
| jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5
| CoyoteAdapter.java
| 
| 
| [EMAIL PROTECTED] wrote:
|  keith   2003/07/02 17:16:49
|  
|Modified:catalina/src/share/org/apache/coyote/tomcat5
|  CoyoteAdapter.java
|Log:
|Allow * as a valid url
| 
| Errr, ok, well, that's a possibly risky patch. I think that does get 
| mapped to the root context and its default servlet (and the servlet path 
| should be *).
| I verified a /* security constraint mapping matched this URL.
| 
| Remy


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



Re: cvs commit: jakarta-tomcat-connectors/jk/tools/reports tomcat_trend.pl

2003-07-03 Thread Glenn Nielsen
Henri Gomez wrote:
[EMAIL PROTECTED] a écrit :

glenn   2003/07/03 07:41:12

  Modified:jk/tools/reports tomcat_trend.pl
  Log:
  Prevent script from failing on bad date format
Revision  ChangesPath
  1.5   +7 -3  
jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl
Index: tomcat_trend.pl
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-connectors/jk/tools/reports/tomcat_trend.pl,v
  retrieving revision 1.4
  retrieving revision 1.5
  diff -u -r1.4 -r1.5
  --- tomcat_trend.pl14 Mar 2003 13:36:58 -1.4
  +++ tomcat_trend.pl3 Jul 2003 14:41:12 -1.5
  @@ -179,6 +179,10 @@
 ($mon, $day, $time, $year) = (split /\s+/,$line)[1..4];
 ($hour, $min, $sec) = split /:/,$time;
 $year =~ s/\]$//;
  +  if( $day !~ /^\d+/ || $hour !~ /^\d+/ || $min!~ /^\d+/ || 
$sec !~ /^\d+/ ) {
  + print Unknown log entry: $origline\n unless $origline =~ 
/\.c /;
  + next;
  +  }
 # print $mon $day $time $year $hour $min $sec\n;
 $logtime = 
timelocal($sec,$min,$hour,$day,$MON{$mon},$year-1900);
   


Will it be included in JK 1.2.5 release ?

Yeah, I'll regenerate the source dist when everyone has finished their
testing and we have a VOTE to do the mod_jk 1.2.5 release.
Glenn

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


Re: [JAKARTA NEWSLETTER DRAFT] News on 'Jakarta Tomcat' from Mayto June, 2003

2003-07-03 Thread Remy Maucherat
Yoav Shapira wrote:
Howdy,
Good paragraph Remy..
Could someone put it in for me (feel free to modify it as needed) ? I've 
always had trouble posting large amounts of data to Nagoya from behind 
my NAT router for some reason (there must be a logical explanation to 
that ...).

P.S. Are here anyone who have been voted in as new committers of Tomcat
within these 2 months?
I've already added myself to your wiki page per your request, and I haven't
checked but I'm sure Tim Funk will get around to it as well ;)
+1 :)

Remy

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


Re: cvs commit: jakarta-tomcat-catalina/catalina/src/share/org/apache/coyote/tomcat5CoyoteAdapter.java

2003-07-03 Thread Remy Maucherat
Keith Wannamaker wrote:
Hi Remy, * does map to the default context.  Can you think of any special 
downstream handling that we should do for '*'?  
Well, nothing as far as I am concerned. I couldn't find any side effect 
or security problem introduced by your patch, so I'm +0 for it.

Remy

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


DO NOT REPLY [Bug 16901] - [PATCH] mod_jk2 wrong server port when using NAT

2003-07-03 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=16901.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=16901

[PATCH] mod_jk2 wrong server port when using NAT





--- Additional Comments From [EMAIL PROTECTED]  2003-07-03 23:14 ---
I just made a similar patch for my mod_jk2 seutp.  Seems pretty important that
jk2 use the Apache Way (TM) of figuring out the server name and port.  The ap_
calls exist for it.

Please apply!

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



DO NOT REPLY [Bug 16901] - [PATCH] mod_jk2 wrong server port when using NAT

2003-07-03 Thread bugzilla
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=16901.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=16901

[PATCH] mod_jk2 wrong server port when using NAT





--- Additional Comments From [EMAIL PROTECTED]  2003-07-03 23:20 ---
FYI, this patch makes jk2 behave in accordance with the UseCanonicalName
documentation:

http://httpd.apache.org/docs-2.0/mod/core.html#usecanonicalname

The current behaviour is unexpected and icky.

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



Re: [JAKARTA NEWSLETTER DRAFT] News on 'Jakarta Tomcat' from May to June, 2003

2003-07-03 Thread Tetsuya Kitahata
On Fri, 04 Jul 2003 00:03:19 +0200
Remy Maucherat [EMAIL PROTECTED] wrote:
 Yoav Shapira wrote:
  Howdy,
  Good paragraph Remy..
 Could someone put it in for me (feel free to modify it as needed) ? I've 
 always had trouble posting large amounts of data to Nagoya from behind 
 my NAT router for some reason (there must be a logical explanation to 
 that ...).

I've already uploaded your nice paragraph to the Apachewiki.
Thank you very much.

Anyone can modify it at
http://nagoya.apache.org/wiki/apachewiki.cgi?JakartaNewsletterDrafts/Issue9

Sincerely,

-- Tetsuya ([EMAIL PROTECTED])


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



Re: SIP Servlet revisited

2003-07-03 Thread Costin Manolache
Brian Olsen wrote:

 Hey Guys,
 
 A little over a year ago the posibility for making an implementation of
 the SIP Servlet specification using Tomcat was shortly talked about
 (see
 http://www.mail-archive.com/[EMAIL PROTECTED]/msg25333.html)
 
 I have done some tests and looked into what problems this would present
 and I came to one conclusion; The current
 design of the org.apache.catalina.Session interface requires that you do
 some weird hacking, twisting and adaptation to get support
 for sessions in SIP, since Session expects to always wrap a HttpSession,
 but SIP Servlets have their own session interface.
 
 I have talked to Anders Kristensen about these difference session
 interfaces. He said that even though there had been talks with
 the Servlet expert group about making a general servlet session
 interface fitting the need of both HTTP and SIP it was voted against
 in the Servlet expert group. The reason they gave for not making such an
 interface was the small spread and use of SIP servlets.
 
 Until such an interface is designed I was wondering if it would be
 possible to change the org.apache.catalina.Session interface to
 return a temporary internal Tomcat interface that contains the same
 methods that would probably end up in the general Servlet
 session interface??
 
 - Brian

Can you explain a bit what you need from the session ? I assume you would
implement the SipRequest, etc - and you could very well implement a sip
session by wrapping the existing session. All you need is extract a session
ID. What is exactly the functionality that is missing ?

Costin


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