cvs commit: jakarta-tomcat-4.0/webapps/admin/WEB-INF controls.tld tree-control.tld

[craigmcc]

craigmcc01/12/17 20:17:32

  Modified:webapps/admin tree-control-test.jsp
  Added:   webapps/admin/WEB-INF controls.tld
  Removed: webapps/admin/WEB-INF tree-control.tld
  Log:
  Change the name of the application-specific tag library to "controls" so
  that we can add some additional gadgets.  No functional changes yet.
  
  Revision  ChangesPath
  1.6   +6 -6  jakarta-tomcat-4.0/webapps/admin/tree-control-test.jsp
  
  Index: tree-control-test.jsp
  ===
  RCS file: /home/cvs/jakarta-tomcat-4.0/webapps/admin/tree-control-test.jsp,v
  retrieving revision 1.5
  retrieving revision 1.6
  diff -u -r1.5 -r1.6
  --- tree-control-test.jsp 2001/11/13 01:00:03 1.5
  +++ tree-control-test.jsp 2001/12/18 04:17:32 1.6
  @@ -4,7 +4,7 @@
   <%@ taglib uri="/WEB-INF/struts-bean.tld" prefix="bean" %>
   <%@ taglib uri="/WEB-INF/struts-html.tld" prefix="html" %>
   <%@ taglib uri="/WEB-INF/struts-logic.tld" prefix="logic" %>
  -<%@ taglib uri="/WEB-INF/tree-control.tld" prefix="tree" %>
  +<%@ taglib uri="/WEB-INF/controls.tld" prefix="controls" %>
   
   
   
  @@ -19,11 +19,11 @@
   
   
   
  -  
   
   
  
  
  
  1.1  jakarta-tomcat-4.0/webapps/admin/WEB-INF/controls.tld
  
  Index: controls.tld
  ===
  
  http://java.sun.com/j2ee/dtds/web-jsptaglibrary_1_2.dtd";>
  
  
  
1.0
1.2
controls

  JSP tag library containing custom GUI controls used in the
  Tomcat Administrative Application.

  

  

  
  tree
  org.apache.webapp.admin.TreeControlTag
  empty
  
Render a "tree" control, based on the current state of a data object
of type org.apache.webapp.admin.TreeControl, which is identified
by the name specified in the "tree" attribute, in the JSP scope
specified by the "scope" attribute.
  
  
  
  action
  false
  true
  
Hyperlink to which expand/contract actions should be sent,
with a string "${node}" marking where the node name of the
affected node should be included (which will usually be as
the value of a request parameter).
  
  
  
  
  images
  false
  true
  
Name of a directory containing the images for our icons,
relative to the page including this tag.  If not specified,
defaults to "images".
  
  
  
  
  scope
  false
  true
  
The JSP scope within which the "tree" attribute is to be found
(page, request, session, or application).  If not specified, the
"tree" attribute will be searched for in any scope.
  
  
  
  
  style
  false
  true
  
The CSS style class to be applied to the entire rendered output
of the tree control.  If not specified, no overall style class
is applied.
  
  
  
  
  styleSelected
  false
  true
  
The CSS style class to be applied to the text of any node that
is currently selected.  If not specified, no style class will be
applied to the text of the selected node.
  
  
  
  
  styleUnselected
  false
  true
  
The CSS style class to be applied to the text of any node that
is *not* currently selected.  If not specified, no style class will
be applied to the text of non-selected nodes.
  
  
  
  
  tree
  false
  true
  
Name of the attribute (in the scope specified by the "scope"
attribute, if any) under which an object of type
org.apache.webapp.admin.TreeControl is stored.  This object
represents the entire current state of the tree, including
a representation of the hierarchical representation of the
nodes, plus the current expanded/ or contracted state of
non-leaf nodes.
  
  
  

  
  
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

cvs commit: jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup Tool.java

[patrickl]

patrickl01/12/17 18:50:30

  Modified:catalina/src/share/org/apache/catalina/startup Tool.java
  Log:
  Implement a better fix for the ArrayIndexOutOfBoundsException in main
  
  Revision  ChangesPath
  1.3   +5 -9  
jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java
  
  Index: Tool.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
  retrieving revision 1.2
  retrieving revision 1.3
  diff -u -r1.2 -r1.3
  --- Tool.java 2001/12/18 02:44:51 1.2
  +++ Tool.java 2001/12/18 02:50:30 1.3
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
 1.2 2001/12/18 02:44:51 patrickl Exp $
  - * $Revision: 1.2 $
  - * $Date: 2001/12/18 02:44:51 $
  + * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
 1.3 2001/12/18 02:50:30 patrickl Exp $
  + * $Revision: 1.3 $
  + * $Date: 2001/12/18 02:50:30 $
*
* 
*
  @@ -116,7 +116,7 @@
* 
*
* @author Craig R. McClanahan
  - * @version $Revision: 1.2 $ $Date: 2001/12/18 02:44:51 $
  + * @version $Revision: 1.3 $ $Date: 2001/12/18 02:50:30 $
*/
   
   public final class Tool {
  @@ -177,14 +177,10 @@
   System.exit(1);
   }
   
  -// Print usage statement if no arguments
  -if (args.length == 0)
  -usage();
  -
   // Process command line options
   int index = 0;
   while (true) {
  -if (index > args.length) {
  +if (index == args.length) {
   usage();
   System.exit(1);
   }
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

cvs commit: jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup Tool.java

[patrickl]

patrickl01/12/17 18:44:51

  Modified:catalina/src/share/org/apache/catalina/startup Tool.java
  Log:
  Print usage statement if there are no arguments
  
  Revision  ChangesPath
  1.2   +8 -4  
jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java
  
  Index: Tool.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
  retrieving revision 1.1
  retrieving revision 1.2
  diff -u -r1.1 -r1.2
  --- Tool.java 2001/11/13 16:57:22 1.1
  +++ Tool.java 2001/12/18 02:44:51 1.2
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
 1.1 2001/11/13 16:57:22 craigmcc Exp $
  - * $Revision: 1.1 $
  - * $Date: 2001/11/13 16:57:22 $
  + * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/startup/Tool.java,v
 1.2 2001/12/18 02:44:51 patrickl Exp $
  + * $Revision: 1.2 $
  + * $Date: 2001/12/18 02:44:51 $
*
* 
*
  @@ -116,7 +116,7 @@
* 
*
* @author Craig R. McClanahan
  - * @version $Revision: 1.1 $ $Date: 2001/11/13 16:57:22 $
  + * @version $Revision: 1.2 $ $Date: 2001/12/18 02:44:51 $
*/
   
   public final class Tool {
  @@ -176,6 +176,10 @@
   log("Must set 'catalina.home' system property");
   System.exit(1);
   }
  +
  +// Print usage statement if no arguments
  +if (args.length == 0)
  +usage();
   
   // Process command line options
   int index = 0;
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5471] New: - jspc -webapp option is broken due to namespace collisions

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5471

jspc -webapp option is broken due to namespace collisions

   Summary: jspc -webapp option is broken due to namespace
collisions
   Product: Tomcat 4
   Version: 4.0.1 Final
  Platform: All
OS/Version: All
Status: NEW
  Severity: Major
  Priority: Other
 Component: Jasper
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


When compiling a web application with the -webapp option, all the Java files 
are written to the same directory and in the same package. If two jsp files, 
such as index.jsp, exist in different directories, they are mapped to the same 
servlet. 

In order to fix this, the command line jsp compiler needs to output the java 
files into directories based on the paths of jsp files, as JspServlet does, and 
also to different packages, which the JspServlet does not. Otherwise the 
different jsp servlets can not be distinguished in web.xml.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5330] - JNDI ENC context problem.

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5330

JNDI ENC context problem.





--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 18:09 ---
Yes, I figured out what different threads are responsible for this tasks:
calls of bound/unbound methods. But other containers, like BEA's Weblogic and 
Orion (these ones I tested)provide access to ENC context in both methods, I 
didn't look at specification yet, but my assumption will be we have to have 
access to ENC context in both cases.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

apx

[Mark Wambach]

I am trying to set up a project which uses servlets and thought I was doing fine until 
it came time to access the server
via port 8180.  All the emails refer to 8080, but it looks to me like the default port 
for non-secure HTTP
communications comes as 8180 in web.xml.

Anyway...

I have jdk1.3.1_01 installed, installed the classes for servlets, and have installed 
Tomcat4.  I'm still able to browse
my website but get no external response from port 8180.  Also, my log files do not 
exist except for the following:

[root@WWW /]# ll /var/tomcat4/logs
total 0
-rw-rw-r--1 root root0 Dec 17 19:54 catalina.out

I have been going nuts trying to compile mod_webapps with -DEAPI and cannot find apx.  
I have the following version of
apache, but did not install it myself:

Server version: Apache/1.3.19 (Unix)  (Red-Hat/Linux)
Server built:   Mar 29 2001 12:52:37

This is related to trying to get servlets to work.  I do not have a browser installed 
on my linux box and only want to
access servlets from outside.  When I added the binary mod_webapps.so to my 
httpd.conf, it gave me the error about
recompiling with -DEAPI.

Secondly, I am curious about the ServerName parameter.  I have it set to localhost.  
Do I need to change it to the IP to
access it from the outside.

I saw someplace that someone recommended just installing another version of Apache.  
Is that the only solution and will
it affect the more recent gcc related files I've installed?

I know I may have committed a multitude of errors, I'd appreciate any pointers.

Thanks for answers to any of these questions.

Mark



--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5463] New: - JKMount directory incorrectly generated in mod_jk.conf for name_number directories under Tomcat default webapps.

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5463

JKMount directory  incorrectly  generated in mod_jk.conf for name_number directories 
under Tomcat default webapps.

   Summary: JKMount directory  incorrectly  generated in mod_jk.conf
for name_number directories under Tomcat default
webapps.
   Product: Tomcat 3
   Version: 3.3 Final
  Platform: PC
OS/Version: Other
Status: NEW
  Severity: Normal
  Priority: Other
 Component: Config
AssignedTo: [EMAIL PROTECTED]
ReportedBy: [EMAIL PROTECTED]


A default directory of name_3 under tomcat webapps generates:

JKMount /name/3/* apj13 -> instead of /name_3/* apj13, etc.

in the auto generated mod_jk.conf. I did not test any generalization of this.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

cvs commit: jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves ErrorReportValve.java

[remm]

remm01/12/17 12:41:00

  Modified:catalina/src/share/org/apache/catalina/valves Tag:
tomcat_40_branch ErrorReportValve.java
  Log:
  - Fix a problem where an exception report thrown from a servlet or a JSP would
not get properly displayed. The bug was caused because sendError now suspends
the response.
  
  Revision  ChangesPath
  No   revision
  
  
  No   revision
  
  
  1.5.2.2   +6 -6  
jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java
  
  Index: ErrorReportValve.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
  retrieving revision 1.5.2.1
  retrieving revision 1.5.2.2
  diff -u -r1.5.2.1 -r1.5.2.2
  --- ErrorReportValve.java 2001/10/04 19:28:13 1.5.2.1
  +++ ErrorReportValve.java 2001/12/17 20:41:00 1.5.2.2
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
 1.5.2.1 2001/10/04 19:28:13 remm Exp $
  - * $Revision: 1.5.2.1 $
  - * $Date: 2001/10/04 19:28:13 $
  + * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
 1.5.2.2 2001/12/17 20:41:00 remm Exp $
  + * $Revision: 1.5.2.2 $
  + * $Date: 2001/12/17 20:41:00 $
*
* 
*
  @@ -104,7 +104,7 @@
* @author Craig R. McClanahan
* @author mailto:[EMAIL PROTECTED]";>Nicola Ken Barozzi Aisa
* @author mailto:[EMAIL PROTECTED]";>Stefano Mazzocchi
  - * @version $Revision: 1.5.2.1 $ $Date: 2001/10/04 19:28:13 $
  + * @version $Revision: 1.5.2.2 $ $Date: 2001/12/17 20:41:00 $
*/
   
   public class ErrorReportValve
  @@ -173,8 +173,6 @@
   if (sresp.isCommitted())
   return;
   
  -response.setSuspended(false);
  -
   ServletRequest sreq = (ServletRequest) request;
   Throwable throwable = 
   (Throwable) sreq.getAttribute(Globals.EXCEPTION_ATTR);
  @@ -197,6 +195,8 @@
   (HttpServletResponse.SC_INTERNAL_SERVER_ERROR);
   
   }
  +
  +response.setSuspended(false);
   
   try {
   report(request, response, throwable);
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

cvs commit: jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves ErrorReportValve.java

[remm]

remm01/12/17 12:21:12

  Modified:catalina/src/share/org/apache/catalina/valves
ErrorReportValve.java
  Log:
  - Fix a problem where an exception report thrown from a servlet or a JSP would
not get properly displayed. The bug was caused because sendError now suspends
the response.
  
  Revision  ChangesPath
  1.7   +13 -12
jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java
  
  Index: ErrorReportValve.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
  retrieving revision 1.6
  retrieving revision 1.7
  diff -u -r1.6 -r1.7
  --- ErrorReportValve.java 2001/10/15 01:11:59 1.6
  +++ ErrorReportValve.java 2001/12/17 20:21:11 1.7
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
 1.6 2001/10/15 01:11:59 remm Exp $
  - * $Revision: 1.6 $
  - * $Date: 2001/10/15 01:11:59 $
  + * $Header: 
/home/cvs/jakarta-tomcat-4.0/catalina/src/share/org/apache/catalina/valves/ErrorReportValve.java,v
 1.7 2001/12/17 20:21:11 remm Exp $
  + * $Revision: 1.7 $
  + * $Date: 2001/12/17 20:21:11 $
*
* 
*
  @@ -104,7 +104,7 @@
* @author Craig R. McClanahan
* @author mailto:[EMAIL PROTECTED]";>Nicola Ken Barozzi Aisa
* @author mailto:[EMAIL PROTECTED]";>Stefano Mazzocchi
  - * @version $Revision: 1.6 $ $Date: 2001/10/15 01:11:59 $
  + * @version $Revision: 1.7 $ $Date: 2001/12/17 20:21:11 $
*/
   
   public class ErrorReportValve
  @@ -169,16 +169,15 @@
   // Perform the request
   context.invokeNext(request, response);
   
  -ServletResponse sresp = (ServletResponse) response;
  -if (sresp.isCommitted())
  -return;
  -
  -response.setSuspended(false);
  -
   ServletRequest sreq = (ServletRequest) request;
   Throwable throwable = 
   (Throwable) sreq.getAttribute(Globals.EXCEPTION_ATTR);
   
  +ServletResponse sresp = (ServletResponse) response;
  +if (sresp.isCommitted()) {
  +return;
  +}
  +
   if (throwable != null) {
   
   // The response is an error
  @@ -198,6 +197,8 @@
   
   }
   
  +response.setSuspended(false);
  +
   try {
   report(request, response, throwable);
   } catch (Throwable tt) {
  @@ -253,7 +254,7 @@
   // Do nothing on an OK status
   if (statusCode == HttpServletResponse.SC_OK)
   return;
  -// Do nothing on a NO MODIFIED status
  +// Do nothing on a NOT MODIFIED status
   if (statusCode == HttpServletResponse.SC_NOT_MODIFIED)
   return;
   // Do nothing on a NO CONTENT status
  @@ -358,7 +359,7 @@
   Writer writer = response.getReporter();
   if (writer != null) {
   // If writer is null, it's an indication that the response has
  -// been hard committed already
  +// been hard committed already, which should never happen
   writer.write(sb.toString());
   writer.flush();
   }
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Re: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[Craig R. McClanahan]

On Mon, 17 Dec 2001, Tom Drake wrote:

>
> Keep in mind that with distributed sessions (coming soon to a tomcat
> near you), affinity is no longer 'required' in order to function. This
> provides the added benefit of fail-over.

As I mentioned in a thread about this on TOMCAT-USER, there *are* session
affinity requirements in the servlet specs (both 2.2 and 2.3) that should
be obeyed.

Craig


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Re: Configuring Multiple Tomcat JVMs with Apache - Load Balancing - Attributes

[Tom Drake]

Oliver:

No, there is no single point of failure. The way it works is thus:

Each tomcat instance in a cluster (defined by a single multicast
address/port):

- Shall keep a copy of all sessions that it owns (created).
- May be configured to be a session repository (or not).
   Session repositories keep copies of all sessions created
   by all members of the cluster.
- Shall automatically discover the other instances at startup time.
   (instances that were started earlier and discovered, shall be
 notified of newly started instances).
- during startup of a new 'repository', will request that it be sent
   copies of all known Sessions (from an arbitrary known repository).
- Shall copy all new / updated sessions to all other known 'repositories'
   after the completion of a web request.
- Shall 'lock' each owned session object while it is in use.
- Shall retrieve and lock a requested session from the 'owner' if the
current
  Tomcat is not the owner of the session.
- If the owner (tomcat) has been brought down in an orderly fashion, it will
   transfer ownership of all of it's owned sessions to arbitrary
repositories
   in the cluster (during it's shutdown sequence).
- If the owner (tomcat) has crashed, the 1st Tomcat (repository) to receive
a request
  for a session that was owned by the downed tomcat will become the new
owner
  of that session.

So,
- There is no single point of failure.
- Server/session affinity, is not required, but will provide performance
benefits.
- New instances may be brought on-line, and will automatically become a part
  of the cluster.
- Existing instances may be brought down with no impact on users (other than
  a one-time performance hit).
- Existing instances may crash with *limited* deleterious side-effects.
- There must be at least two 'repository' instances in order to support the
   fail-over scenarios. However, the number of 'repository' instances should
   be limited to as small a number as is necessary to meet your fail-over
needs.
- This feature will only be present (and consume bandwidth) if it has been
   configured. Other load balancing solutions that don't require fail-over
   protection, such as the Tomcat 3.x / Apache solution can be implemented
   without concern for this.

- Original Message -
From: "Lauer, Oliver" <[EMAIL PROTECTED]>
To: "'Tomcat Developers List'" <[EMAIL PROTECTED]>; "'Tom
Drake'" <[EMAIL PROTECTED]>
Sent: Monday, December 17, 2001 9:25 AM
Subject: AW: Configuring Multiple Tomcat JVMs with Apache - Load Balancing


| Tom,
|
| Will the repository be a single point of failure then ?
| Oliver
|
| > AXA eSolutions GmbH
| > AXA Konzern AG Germany
| > Oliver Lauer
| > Web Architect
| > Wörthstraße 34
| > D-50668 Köln
| > Germany
| > Tel.: +49 221 148 31277
| > Fax: +49 221 148 43963
| > Mobil: +49 179 59 064 59
| > e-Mail: [EMAIL PROTECTED]
| > _
| >
|
|
| -Ursprüngliche Nachricht-
| Von: Tom Drake [mailto:[EMAIL PROTECTED]]
| Gesendet: Montag, 17. Dezember 2001 18:26
| An: Tomcat Developers List
| Betreff: Re: Configuring Multiple Tomcat JVMs with Apache - Load
| Balancing
|
|
| Keep in mind that with distributed sessions (coming soon to a tomcat near
| you), affinity
| is no longer 'required' in order to function. This provides the added
| benefit of fail-over.
|
| Affinity is still desirable for performance reasons. However, the if the
| Tomcat instance
| that created the session is brought down, the user needn't be affected.
| Apache
| should notice this and route subsequent requests (for the affected
sessions)
| to another
| TC instance in the cluster. The other TC instances will either already
have
| a copy of
| the session or can retrieve it from a 'repository'.
|
| Tom
|
| - Original Message -
| From: <[EMAIL PROTECTED]>
| To: "Tomcat Developers List" <[EMAIL PROTECTED]>
| Sent: Monday, December 17, 2001 9:07 AM
| Subject: RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing
|
|
| | On Mon, 17 Dec 2001, GOMEZ Henri wrote:
| |
| | > >couldn't the connector be solely
| | > >responsible for
| | > >this?  the connector could add a cookie to the response, then read
this
| | > >cookie on subsequent requests, and then you wouldn't have to
| | > >add stuff to
| | > >servlet containers to handle this.  or, i could be totally off base
:)
| | >
| | > Could we add another cookie in http reply ? Spec compliant ?
| | > What to be used when you have to use URL encoded instead since
| | > cookies are not available ?
| |
| | It'll not work. URL rewriting is the issue - the session id ( i.e. what
| | will be sent on the next request ) will end up inside html pages, where
| | mod_jk can't 'rewrite' it. And adding a cookie will fail if no cookies
| | are available ( again where url rewriting will be used ).
| |
| | The session id that a servlet sees must have the vmroute in it.
| |
| |
| | Costin
| |
| |
| | --
| | To unsubscribe, e-mail:
| 
| | For additional comman

cvs commit: jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin SetUpServerAction.java

[patrickl]

patrickl01/12/17 09:37:43

  Modified:webapps/admin/WEB-INF/classes/org/apache/webapp/admin
SetUpServerAction.java
  Log:
  Replace tabs with proper number of spaces to conform to coding style standard
  
  Revision  ChangesPath
  1.5   +27 -27
jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java
  
  Index: SetUpServerAction.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java,v
  retrieving revision 1.4
  retrieving revision 1.5
  diff -u -r1.4 -r1.5
  --- SetUpServerAction.java2001/12/17 17:34:05 1.4
  +++ SetUpServerAction.java2001/12/17 17:37:43 1.5
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java,v
 1.4 2001/12/17 17:34:05 patrickl Exp $
  - * $Revision: 1.4 $
  - * $Date: 2001/12/17 17:34:05 $
  + * $Header: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java,v
 1.5 2001/12/17 17:37:43 patrickl Exp $
  + * $Revision: 1.5 $
  + * $Date: 2001/12/17 17:37:43 $
*
* 
*
  @@ -98,7 +98,7 @@
*
* @author Jazmin Jonson
* @author Manveen Kaur
  - * @version $Revision: 1.4 $ $Date: 2001/12/17 17:34:05 $
  + * @version $Revision: 1.5 $ $Date: 2001/12/17 17:37:43 $
*/
   
   public class SetUpServerAction extends Action {
  @@ -137,27 +137,27 @@
   
   HttpSession session = request.getSession();
   
  -   // Do I have to do this part ??
  - if (form == null) {
  -getServlet().log(" Creating new ServerForm bean under key "
  -+ mapping.getAttribute());
  - form = new ServerForm();
  -
  +// Do I have to do this part ??
  +if (form == null) {
  +getServlet().log(" Creating new ServerForm bean under key "
  ++ mapping.getAttribute());
  +form = new ServerForm();
  +   
   if ("request".equals(mapping.getScope()))
   request.setAttribute(mapping.getAttribute(), form);
   else
   request.getSession().setAttribute(mapping.getAttribute(), form);
  -
  - }
  +   
  +}
   
   MessageResources messages = getResources();
   Locale locale = (Locale)session.getAttribute(Action.LOCALE_KEY);
   
  - // Do transaction stuff before this
  +// Do transaction stuff before this
   
  - ServerForm serverFm = (ServerForm) form;
  +ServerForm serverFm = (ServerForm) form;
   
  - if(debugLvlList == null) {
  +if(debugLvlList == null) {
   debugLvlList = new ArrayList();
   debugLvlList.add(new LabelValueBean("0", "0"));
   debugLvlList.add(new LabelValueBean("1", "1"));
  @@ -172,10 +172,10 @@
   
   }
   
  - if(actionList == null) {
  +if(actionList == null) {
   
   actionList = new ArrayList();
  -   // You can get this from the Mbean
  +// You can get this from the Mbean
   actionList.add(new LabelValueBean(
   messages.getMessage(locale, "actions.available.actions"),
   "Available Actions"));
  @@ -189,15 +189,15 @@
   "Delete Services"));
   }
   
  - Integer portNumb = null;
  - Integer debug = null;
  - String shutdown = null;
  +Integer portNumb = null;
  +Integer debug = null;
  +String shutdown = null;
   try{
  - 
  +
   if(mBServer == null) {
  -ApplicationServlet servlet = (ApplicationServlet)getServlet();  
 
  +ApplicationServlet servlet = (ApplicationServlet)getServlet();  
  
   mBServer = servlet.getServer();
  - }
  +}
   
   Iterator serverItr = 
   mBServer.queryMBeans(new ObjectName(TomcatTreeBuilder.SERVER_TYPE +
  @@ -207,16 +207,16 @@
   ObjectName serverObjName =
   ((ObjectInstance)serverItr.next()).getObjectName();
   
  - /*
  +/*
   ModelMBeanInfo info = (ModelMBeanInfo) 
mBServer.getMBeanInfo(serverObjName);  
   MBeanAttributeInfo attrs[] = info.getAttributes();
   for (int i = 0; i < attrs.length; i++)
   System.out.println("  AttributeInfo=" + attrs[i]);
  - 
  +
   MBeanOperationInfo opers[] = info.getOperations();
   for (int i = 0; i < opers.length; i++)
   System.out.println("  Operation=" + opers[i]);
  - */
  +   */
   
  

cvs commit: jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin ApplicationResources_en.properties ApplicationResources_es.properties SetUpServerAction.java SetUpServiceAction.java

[patrickl]

patrickl01/12/17 09:34:05

  Modified:webapps/admin/WEB-INF/classes/org/apache/webapp/admin
ApplicationResources_en.properties
ApplicationResources_es.properties
SetUpServerAction.java SetUpServiceAction.java
  Log:
  Correct the following localization problems:
  - Make the accented characters conform to the Latin-1 Supplement chart at the
following URL so that they appear correctly on all browsers and platforms:
 http://www.unicode.org/charts/PDF/U0080.pdf
  - Add the user's current locale as an argument to MessageResources.getMessage()
so that the correct message is loaded instead of the default message.
  
  Revision  ChangesPath
  1.9   +1 -1  
jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/ApplicationResources_en.properties
  
  Index: ApplicationResources_en.properties
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/ApplicationResources_en.properties,v
  retrieving revision 1.8
  retrieving revision 1.9
  diff -u -r1.8 -r1.9
  --- ApplicationResources_en.properties2001/12/16 19:40:20 1.8
  +++ ApplicationResources_en.properties2001/12/17 17:34:05 1.9
  @@ -38,7 +38,7 @@
   service.properties=Service Properties
   service.property=Property
   service.value=Value
  -actions.avaliable.actions=Available Actions
  +actions.available.actions=Available Actions
   actions.services.create=Create New Service
   actions.services.delete=Delete Services
   actions.accesslogger.create=Create New Access Logger
  
  
  
  1.11  +13 -13
jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/ApplicationResources_es.properties
  
  Index: ApplicationResources_es.properties
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/ApplicationResources_es.properties,v
  retrieving revision 1.10
  retrieving revision 1.11
  diff -u -r1.10 -r1.11
  --- ApplicationResources_es.properties2001/12/16 19:40:20 1.10
  +++ ApplicationResources_es.properties2001/12/17 17:34:05 1.11
  @@ -38,21 +38,21 @@
   service.properties=Mantenga Las Caracteristicas
   service.property=Caracteristica
   service.value=Valor
  -actions.avaliable.actions=Disponibles Acciones
  +actions.available.actions=Disponibles Acciones
   actions.services.create=Cree El Nuevo Servicio
  -actions.services.delete=Servicios De la Cancelaci\u00dbn
  -actions.accesslogger.create=Cree El Maderero Nuevo Del Acceso
  -actions.accesslogger.delete=Maderero Del Acceso De la Cancelaci\u00dbn
  +actions.services.delete=Servicios de la Cancelaci\u00f3n
  +actions.accesslogger.create=Cree El Maderero Nuevo del Acceso
  +actions.accesslogger.delete=Maderero del Acceso de la Cancelaci\u00f3n
   actions.connector.create=Cree Nuevo El Conector
  -actions.connector.delete=Conectores De la Cancelaci\u00dbn
  +actions.connector.delete=Conectores de la Cancelaci\u00f3n
   actions.host.create=Cree El Nuevo Ordenador principal
  -actions.host.delete=Ordenadores principal De la Cancelaci\u00dbn
  +actions.host.delete=Ordenadores principal de la Cancelaci\u00f3n
   actions.logger.create=Cree Nuevo El Maderero
  -actions.logger.delete=Madereros De la Cancelaci\u00dbn
  -actions.requestfilter.create=Cree El Filtro Nuevo De la Petici\u00dbn
  -actions.requestfilter.delete=Filtros De la Petici\u00dbn De la Cancelaci\u00dbn
  -actions.userrealm.create=Cree El Nuevo Reino Del Utilizador
  -actions.userrealm.delete=Reinos Del Utilizador De la Cancelaci\u00dbn
  -actions.valve.create=Cree Nueva La V\u00b7lvula
  -actions.valve.delete=V\u00b7lvulas De la Cancelaci\u00dbn
  +actions.logger.delete=Madereros de la Cancelaci\u00f3n
  +actions.requestfilter.create=Cree El Filtro Nuevo de la Petici\u00f3n
  +actions.requestfilter.delete=Filtros de la Petici\u00f3n de la Cancelaci\u00f3n
  +actions.userrealm.create=Cree El Nuevo Reino del Utilizador
  +actions.userrealm.delete=Reinos del Utilizador de la Cancelaci\u00f3n
  +actions.valve.create=Cree Nueva La V\u00e1lvula
  +actions.valve.delete=V\u00e1lvulas de la Cancelaci\u00f3n
   actions.service.delete=Suprima este servicio
  
  
  
  1.4   +10 -8 
jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java
  
  Index: SetUpServerAction.java
  ===
  RCS file: 
/home/cvs/jakarta-tomcat-4.0/webapps/admin/WEB-INF/classes/org/apache/webapp/admin/SetUpServerAction.java,v
  retrieving revision 1.3
  retrieving revision 1.4
  diff -u -r1.3 -r1.4
  --- SetUpServerAction.java2001/12/16 19:40:20 1.3
  +++ SetUpServerAction.java2001/12/17 17:34:05 1.4
  @@ -1,7 +1,7 @@
   /*
  - * $Header: 
/home/cvs/jakarta-tom

AW: Configuring Multiple Tomcat JVMs with Apache - LoadBalancing

[Lauer, Oliver]

Tom, 

Will the repository be a single point of failure then ?
Oliver  

> AXA eSolutions GmbH
> AXA Konzern AG Germany
> Oliver Lauer 
> Web Architect
> Wörthstraße 34
> D-50668 Köln
> Germany
> Tel.: +49 221 148 31277
> Fax: +49 221 148 43963
> Mobil: +49 179 59 064 59
> e-Mail: [EMAIL PROTECTED]
> _
> 


-Ursprüngliche Nachricht-
Von: Tom Drake [mailto:[EMAIL PROTECTED]]
Gesendet: Montag, 17. Dezember 2001 18:26
An: Tomcat Developers List
Betreff: Re: Configuring Multiple Tomcat JVMs with Apache - Load
Balancing


Keep in mind that with distributed sessions (coming soon to a tomcat near
you), affinity
is no longer 'required' in order to function. This provides the added
benefit of fail-over.

Affinity is still desirable for performance reasons. However, the if the
Tomcat instance
that created the session is brought down, the user needn't be affected.
Apache
should notice this and route subsequent requests (for the affected sessions)
to another
TC instance in the cluster. The other TC instances will either already have
a copy of
the session or can retrieve it from a 'repository'.

Tom

- Original Message -
From: <[EMAIL PROTECTED]>
To: "Tomcat Developers List" <[EMAIL PROTECTED]>
Sent: Monday, December 17, 2001 9:07 AM
Subject: RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing


| On Mon, 17 Dec 2001, GOMEZ Henri wrote:
|
| > >couldn't the connector be solely
| > >responsible for
| > >this?  the connector could add a cookie to the response, then read this
| > >cookie on subsequent requests, and then you wouldn't have to
| > >add stuff to
| > >servlet containers to handle this.  or, i could be totally off base :)
| >
| > Could we add another cookie in http reply ? Spec compliant ?
| > What to be used when you have to use URL encoded instead since
| > cookies are not available ?
|
| It'll not work. URL rewriting is the issue - the session id ( i.e. what
| will be sent on the next request ) will end up inside html pages, where
| mod_jk can't 'rewrite' it. And adding a cookie will fail if no cookies
| are available ( again where url rewriting will be used ).
|
| The session id that a servlet sees must have the vmroute in it.
|
|
| Costin
|
|
| --
| To unsubscribe, e-mail:

| For additional commands, e-mail:

|
|
|


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 


--
Aus Rechts- und Sicherheitsgruenden ist die in dieser E-Mail gegebene Information 
nicht rechtsverbindlich. Eine rechtsverbindliche Bestaetigung reichen wir Ihnen gerne 
auf Anforderung in schriftlicher Form nach. Beachten Sie bitte, dass jede Form der 
unautorisierten Nutzung, Veroeffentlichung, Vervielfaeltigung oder Weitergabe des 
Inhalts dieser E-Mail nicht gestattet ist.Diese Nachricht  ist ausschliesslich fuer 
den bezeichneten Adressaten oder dessen Vertreter bestimmt. Sollten Sie nicht der 
vorgesehene Adressat dieser E-Mail oder dessen Vertreter sein, so bitten wir Sie, sich 
mit dem Absender der E-Mail in Verbindung zu setzen.

For legal and security reasons the information provided in this e-mail is not legally 
binding. Upon request we would be pleased to provide you with a legally binding 
confirmation in written form. Any form of unauthorised use, publication, reproduction, 
copying or disclosure of the content of this e-mail is not permitted. This message is 
exclusively for the person addressed or their representative. If you are not the 
intended recipient of this message and its contents, please notify the sender 
immediately.

==


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[Craig R. McClanahan]

On Mon, 17 Dec 2001, Kevin Seguin wrote:

> Date: Mon, 17 Dec 2001 10:25:17 -0600
> From: Kevin Seguin <[EMAIL PROTECTED]>
> Reply-To: Tomcat Developers List <[EMAIL PROTECTED]>
> To: 'Tomcat Developers List' <[EMAIL PROTECTED]>
> Subject: RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing
>
> >
> > >there hasn't been done anything on that topic yet ?  What's
> > >the status of
> > >loadbalacing, either mod_jk or mod_webapp ?
> > >Is that political due to if loadbalacing is working properly
> > >there won't be
> > >any reason to take (buy) anything else than TC ?
> >
> > State of the art is that today only mod_jk could
> > handle load-balancing and only when connected
> > to Tomcat 3.2.x or 3.3.
> >
> > Tomcat 4.0.x support ajp13 protocol, used in mod_jk
> > but still miss a subtil feature (jvmroute) to be
> > able to keep the route to the good JVM in
> > session mode (SessionAfinity).
> >
> > But the current refactory of ajp protocol,
> > under ajp14 in jakarta-tomcat-connectors, will
> > fix somedays thanks to Costin and Kevin works :)))
> >
>
> how does the connector set/use jvmroute?  i assume there is some sort of
> cookie magic going on, but after a brief tour of the code, it's not obvious
> to me how this jvmroute stuff works.
>

That was also my impression when I looked.

> it seems to me that tomcat (any version) shouldn't have to have anything to
> do with setting jvmroute.  couldn't the connector be solely responsible for
> this?  the connector could add a cookie to the response, then read this
> cookie on subsequent requests, and then you wouldn't have to add stuff to
> servlet containers to handle this.  or, i could be totally off base :)
>

If someone familiar with how "jvmroute" works wanted to post some patches
for Tomcat 4, we could have it working with the AJPv13 method.  (Or, if
you're a committer, you have commit access on "jakarta-tomcat-4.0" as well
:-).

> -kevin.
>

Craig


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[costinm]

On Mon, 17 Dec 2001, GOMEZ Henri wrote:

> >couldn't the connector be solely
> >responsible for
> >this?  the connector could add a cookie to the response, then read this
> >cookie on subsequent requests, and then you wouldn't have to
> >add stuff to
> >servlet containers to handle this.  or, i could be totally off base :)
>
> Could we add another cookie in http reply ? Spec compliant ?
> What to be used when you have to use URL encoded instead since
> cookies are not available ?

It'll not work. URL rewriting is the issue - the session id ( i.e. what
will be sent on the next request ) will end up inside html pages, where
mod_jk can't 'rewrite' it. And adding a cookie will fail if no cookies
are available ( again where url rewriting will be used ).

The session id that a servlet sees must have the vmroute in it.


Costin


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[costinm]

On Mon, 17 Dec 2001, Kevin Seguin wrote:

> > Tomcat 4.0.x support ajp13 protocol, used in mod_jk
> > but still miss a subtil feature (jvmroute) to be
> > able to keep the route to the good JVM in
> > session mode (SessionAfinity).
>
> how does the connector set/use jvmroute?  i assume there is some sort of
> cookie magic going on, but after a brief tour of the code, it's not obvious
> to me how this jvmroute stuff works.

That's quite simple. Each session generated by tomcat has a prefix
containing the 'worker id'.

mod_jk will look at all incoming sessions, and if it detects the prefix
it'll quickly send the request directly to the right worker ( the
'afinity' ).

Now the only 'trick' is to get the tomcat session manager to generate the
session id including the 'worker id' ( i.e. ajp13, or ajp13_9009, or
whatever you set in your workers.properties ).

mod_jk will send the worker id on each request, so tomcat will know how it
was received and pass this info to the session manager when it generates
the session id.

With 4.0, the Manager doesn't have access to the Request object, so the
only solution I found so far is to add a property that is set from
server.xml settings ( that's quite trivial to add ). The user will
have to keep workers.properties and server.xml in sync, i.e. use the
same worker name ( and workerId property ).


> it seems to me that tomcat (any version) shouldn't have to have anything to
> do with setting jvmroute.  couldn't the connector be solely responsible for
> this?  the connector could add a cookie to the response, then read this

That's another solution ( I tryed to do this ), but it'll not work.

The 'session id' is visible to servlets, etc. A servlet may expect
a cookie ( or url ) with the session id. URL rewriting may generate
urls with the session id in it.

Mod_jk could rewrite the cookie on the fly, but can't go into
html and change the urls.

So whoever generates the session id must include the route ( == worker
id). I don't think there's any other way - mod_jk doesn't have
control over that ( well, it has in 3.3, using the new session hook, but
it's too confusing to be a good solution ).

Costin


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[GOMEZ Henri]

>how does the connector set/use jvmroute?  i assume there is 
>some sort of
>cookie magic going on, but after a brief tour of the code, 
>it's not obvious
>to me how this jvmroute stuff works.

the jvmroute is computed when sessionid is created :

sessionid = f(entropy) + jvmroute

>it seems to me that tomcat (any version) shouldn't have to 
>have anything to
>do with setting jvmroute.  

if tomcat didn't set the jvmroute, you'll have to determine
on connector :

1) that a session has been created and get the sessionid, which
   is hidden into http reply headers (at least in ajp12/ajp13)

2) Associate that session id with the ajp worker and keep it.

3) Each time you get a request, you have to get the worker for
   that session and route the request to it

4) You also have to handle the session timeout on connector
   side to remove then when outdated, and there you could be
   stucked (may be not today with Costin trick of sending 
   http request to web-server ;)
 
>couldn't the connector be solely 
>responsible for
>this?  the connector could add a cookie to the response, then read this
>cookie on subsequent requests, and then you wouldn't have to 
>add stuff to
>servlet containers to handle this.  or, i could be totally off base :)

Could we add another cookie in http reply ? Spec compliant ?
What to be used when you have to use URL encoded instead since
cookies are not available ?

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Transfer-Encoding: chunked

[Will Droste]

Hey Guys,

My name is Will I've been trying to do Transfer-Encoding: chunked on my
request to Tomcat 4.0.1, with no success I've tried using the Apache Commons
jar, a HttpClient library that said it had chunked encoding on requests, and
my own implementation.  I've look through the Catalina source tree and
noticed there was code for chunked encoding but I was unable to diagnosis
the problem.  Any help would be greatly appricated.  I was using RFC 2616 as
a reference for my implementation of HTTP/1.1.  If you have some test code
that I could look at the proves that it work that would be awesome.

Regards,
Will

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[Kevin Seguin]

> 
> >there hasn't been done anything on that topic yet ?  What's 
> >the status of
> >loadbalacing, either mod_jk or mod_webapp ? 
> >Is that political due to if loadbalacing is working properly 
> >there won't be
> >any reason to take (buy) anything else than TC ?  
> 
> State of the art is that today only mod_jk could 
> handle load-balancing and only when connected 
> to Tomcat 3.2.x or 3.3.
> 
> Tomcat 4.0.x support ajp13 protocol, used in mod_jk
> but still miss a subtil feature (jvmroute) to be
> able to keep the route to the good JVM in
> session mode (SessionAfinity).
> 
> But the current refactory of ajp protocol,
> under ajp14 in jakarta-tomcat-connectors, will
> fix somedays thanks to Costin and Kevin works :)))
> 

how does the connector set/use jvmroute?  i assume there is some sort of
cookie magic going on, but after a brief tour of the code, it's not obvious
to me how this jvmroute stuff works.

it seems to me that tomcat (any version) shouldn't have to have anything to
do with setting jvmroute.  couldn't the connector be solely responsible for
this?  the connector could add a cookie to the response, then read this
cookie on subsequent requests, and then you wouldn't have to add stuff to
servlet containers to handle this.  or, i could be totally off base :)

-kevin.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Fw: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[Tom Drake]

- Original Message -
From: "Tom Drake" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, December 17, 2001 7:47 AM
Subject: Re: Configuring Multiple Tomcat JVMs with Apache - Load Balancing


| When my work is done, Session Affinity will not be required in order to
| acheive load balancing. However, we should still want to use session
| affinity in our load balancing solution(s) for the simple reason that
| it will perform better.
|
| However, the session affinity between Apache and Tomcat 3 locks
| a user (based on her JSESSIONID) to a single Tomcat instance. Once
| the distributed session management solution is in place, Apache should
| 'prefer' NOT 'force' tomcat instance routing. This gives us a real
fail-over
| story. Administrators will be able to bring down Tomcat instances without
| blowing away 'logged in' users.
|
| Tom
|
| - Original Message -
| From: "GOMEZ Henri" <[EMAIL PROTECTED]>
| To: "Tomcat Developers List" <[EMAIL PROTECTED]>
| Cc: "Tomcat Users List (E-Mail)" <[EMAIL PROTECTED]>
| Sent: Monday, December 17, 2001 12:54 AM
| Subject: RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing
|
|
| | >there hasn't been done anything on that topic yet ?  What's
| | >the status of
| | >loadbalacing, either mod_jk or mod_webapp ?
| | >Is that political due to if loadbalacing is working properly
| | >there won't be
| | >any reason to take (buy) anything else than TC ?
| |
| | State of the art is that today only mod_jk could
| | handle load-balancing and only when connected
| | to Tomcat 3.2.x or 3.3.
| |
| | Tomcat 4.0.x support ajp13 protocol, used in mod_jk
| | but still miss a subtil feature (jvmroute) to be
| | able to keep the route to the good JVM in
| | session mode (SessionAfinity).
| |
| | But the current refactory of ajp protocol,
| | under ajp14 in jakarta-tomcat-connectors, will
| | fix somedays thanks to Costin and Kevin works :)))
| |
| | --
| | To unsubscribe:   
| | For additional commands: 
| | Troubles with the list: 
| |
| |
| |
|
|
| --
| To unsubscribe:   
| For additional commands: 
| Troubles with the list: 
|
|
|


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: [j-t-c] patch for jakarta-tomcat-connectors/jk/native/common/ jk_a jp_common.c

[Kevin Seguin]

> 
> The patch seems valid and the pmsg was passed for this
> purpose in previous release of J-T-C but some folks
> (didn't remember who) make modifications in mod_jk
> part of code and the update was lost.
> 
> Seems ok, you could commit :)
> 

done.

> This patch fix a real problem and should be also 
> propagated to Tomcat 3.3.1 mod_jk 
> 

i looked at the source on the HEAD of jakarta-tomcat/src/native/mod_jk and
didn't see a similar problem.  it's quite possible that i missed something
:)

-kevin.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4223] - AJP13 connectors InputStream newer returns -1

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4223

AJP13 connectors InputStream newer returns -1

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||FIXED



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 07:31 ---
this bug was due to a bug in jk, as suspected.

it is now fixed on the HEAD of jakarta-tomcat-connectors.  the fix was in 
jk/native/common/jk_ajp_common.c.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

cvs commit: jakarta-tomcat-connectors/jk/native/common jk_ajp_common.c

[seguin]

seguin  01/12/17 07:29:38

  Modified:jk/native/common jk_ajp_common.c
  Log:
  actually send post data beyond the first ~8K to the servlet container.
  
  if a client request contains post data, the first chunk of it is sent to the
  container directly after the forward request is sent.  if not all of the data
  fits in the first chunk, the container will get subsequent chunks by
  sending the connector a GET_BODY_CHUNK message.  this checkin fixes the
  reponse to the GET_BODY_CHUNK message so that the connector will sent the
  container more post data.
  
  i believe this also fixes bug 4223.
  PR: 4223
  
  Revision  ChangesPath
  1.22  +2 -2  jakarta-tomcat-connectors/jk/native/common/jk_ajp_common.c
  
  Index: jk_ajp_common.c
  ===
  RCS file: /home/cvs/jakarta-tomcat-connectors/jk/native/common/jk_ajp_common.c,v
  retrieving revision 1.21
  retrieving revision 1.22
  diff -u -r1.21 -r1.22
  --- jk_ajp_common.c   2001/12/04 19:44:23 1.21
  +++ jk_ajp_common.c   2001/12/17 15:29:38 1.22
  @@ -59,7 +59,7 @@
* Description: common stuff for bi-directional protocols ajp13/ajp14. *
* Author:  Gal Shachor <[EMAIL PROTECTED]>   *
* Author:  Henri Gomez <[EMAIL PROTECTED]>   *
  - * Version: $Revision: 1.21 $   *
  + * Version: $Revision: 1.22 $   *
***/
   
   
  @@ -942,7 +942,7 @@
}
   
/* the right place to add file storage for upload */
  - if ((len = ajp_read_into_msg_buff(ae, r, msg, len, l)) >= 0) {
  + if ((len = ajp_read_into_msg_buff(ae, r, pmsg, len, l)) >= 0) {
r->content_read += len;
return JK_AJP13_HAS_RESPONSE;
}  
  
  
  

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: org.apache.tomcat.util.net package in tomcat 3.x

[GOMEZ Henri]

>If the trafic becomes too big, we should definitely separate them, but
>what is big remain open ( tomcat-users is huge, I just can't read it
>without headaches - that's where a split will be most needed )

Exact, we'll have to split at some time just to be able to 
see if the question is related to core or connector land ;)


--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5446] - Can't change webapp class loader (bug #5391 revisited :)

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5446

Can't change webapp class loader (bug #5391 revisited :)





--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 02:28 ---
One fix to my earlier note: instanceof is not a proper way to check it, I
suppose. But something like:

Class clazz = Class.forName(loaderClass);
if (clazz.isAssignableFrom(WebappClassLoader.class)) {
   classLoader = (WebappClassLoader) clazz.newInstance();
  ..
}
else {
   throw Exception("configuration error")
}

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5446] - Can't change webapp class loader (bug #5391 revisited :)

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5446

Can't change webapp class loader (bug #5391 revisited :)

[EMAIL PROTECTED] changed:

   What|Removed |Added

   Severity|Normal  |Minor
 Status|RESOLVED|REOPENED
 Resolution|WONTFIX |



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 02:17 ---
Ok, the use case of using another CL which would cast to WebappCL is valid. I
suppose replacing the new WebappClassloader by a Class.forName.newInstance (if
the loaderClass is defined) would work.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5446] - Can't change webapp class loader (bug #5391 revisited :)

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5446

Can't change webapp class loader (bug #5391 revisited :)





--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 02:10 ---
Thanks for the comments! In my case I would very much like to reuse the already
implemented features of WebappLoader since my own class loader inherits
WebappClassLoader and only overwrites loadClass method. That is, it is also a
WebappClassLoader. Now that WebappLoader and WebappClassLoader are tightly
integrated with a specialized API, would it be possible to support this feature
in case the replacing class loader extends WebappClassLoader? (i.e.
MyClassLoader instanceof WebappClassLoader == true). 

This feature would be extremely nice and it wouldn't break the protocol between
WebappLoader and WebappClassLoader :)

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5379] - j-t-c webapps java can not compile

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5379

j-t-c webapps java can not compile

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|AJP Connector   |WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4139] - mod_webapp build files for Win32

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4139

mod_webapp build files for Win32

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4042] - webapp component requires Port directive versus Listen directive

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4042

webapp component requires Port directive versus Listen directive

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 3643] - Enable static-linking within Apache for WebApp module

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=3643

Enable static-linking within Apache for WebApp module

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4500] - isapi_redirect.dll does not pass Client certificate to tomcat

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4500

isapi_redirect.dll does not pass Client certificate to tomcat

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|AJP Connector
Summary|isapi_redirect.dll does not |isapi_redirect.dll does not
   |pass Client certificate to  |pass Client certificate to
   |tomcat  |tomcat

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5446] - Can't change webapp class loader (bug #5391 revisited :)

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5446

Can't change webapp class loader (bug #5391 revisited :)

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||WONTFIX



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 01:40 ---
Although the API may indicate it's doable, it's not possible to separate the
Loader from the ClassLoader in the case of the WebappLoader/WebappClassLoader.
The first one uses special APIs from the second one that can't be replaced.
If you want to replace the class loader, you should write an alternate loader
(maybe based on the old StandardLoader), and maybe give it the capabilities to
switch the CL.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4223] - AJP13 connectors InputStream newer returns -1

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4223

AJP13 connectors InputStream newer returns -1

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|AJP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4930] - java.io.StreamCorruptedException: Type code out of range, is 0 with Apache WebApp module

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4930

java.io.StreamCorruptedException: Type code out of range, is 0 with Apache WebApp 
module

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 3509] - Apache 1.3.20 & mod_webapp & Tomcat 4b7 HANGS under Win

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=3509

Apache 1.3.20 & mod_webapp & Tomcat 4b7 HANGS under Win

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Re: Trouble building mod_webapp.so for Apache 2.0.28

[jean-frederic clere]

Try to update your libtool.
I use the following:
+++
jfclere@vtxclere:~ > libtool --version
ltmain.sh (GNU libtool) 1.3.5 (1.385.2.206 2000/05/27 11:12:27)
+++

Cavan Morris wrote:
> 
> I am unable to build mod_webapp.so for apache 2.0.28 on my redhat 7.2
> system.
> 
> Here's what I've been doing.  The problem comes during the make process.
> 
> > tar -xvzf webapp-module-1.0.1-tc401-src.tar.gz
> [...]
> > cd webapp-module-1.0.1-tc401/
> [...]
> > ./support/buildconf.sh
> [...]
> > ./configure --with-apxs=/usr/local/apache2/bin/apxs
> [...]
> > make
> [...]
> *** Warning: This library needs some functionality provided by -lwebapp.
> *** I have the capability to make that library automatically link in
> when
> *** you link to this library.  But I can only do this if you have a
> *** shared version of the library, which you do not appear to have.
> 
> *** Warning: libtool could not satisfy all declared inter-library
> *** dependencies of module mod_webapp.  Therefore, libtool will create
> *** a static module, that should work as long as the dlopening
> *** application is linked with the -dlopen flag.
> make[2]: Leaving directory `/root/webapp-module-1.0.1-tc401/apache-2.0'
> make[1]: Exiting directory "apache-2.0"
> make[1]: Leaving directory `/root/webapp-module-1.0.1-tc401'
> 
> At this point make finishes and there is no mod_webapp.so in apache-2.0.
> Does anyone have any idea how to get this to work?  Any help would be
> greatly appriciated.
> 
> Thanks in advance.
> Cavan Morris
> [EMAIL PROTECTED]
> 
> --
> To unsubscribe, e-mail:   
> For additional commands, e-mail: 

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5364] - Can't re-deploy a web application with the manager

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5364

Can't re-deploy a web application with the manager

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||WONTFIX



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 01:35 ---
As stated by Craig and others (me included), the JDK has some issues with
reading updated zip files. That causes problems here, and that also causes
problems for class reloading (when you update a JAR in /WEB-INF/lib).
Unfortunately, it seems it can't be fixed unless the JDK is fixed (or we use an
alternate library to read the zip), at which point it should start working. I
don't know if it would work better with an IBM JDK.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4334] - ServerName in httpd.conf vs. defaultHost in server.xml Error: addChild: Child name 'hostname.myserver.com' is not unique

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4334

ServerName in httpd.conf vs. defaultHost in server.xml Error: addChild:  Child name 
'hostname.myserver.com' is not unique

[EMAIL PROTECTED] changed:

   What|Removed |Added

  Component|Other Connectors|WARP Connector

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 4334] - ServerName in httpd.conf vs. defaultHost in server.xml Error: addChild: Child name 'hostname.myserver.com' is not unique

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=4334

ServerName in httpd.conf vs. defaultHost in server.xml Error: addChild:  Child name 
'hostname.myserver.com' is not unique

[EMAIL PROTECTED] changed:

   What|Removed |Added

 CC||[EMAIL PROTECTED]



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 01:30 
---
*** Bug 5438 has been marked as a duplicate of this bug. ***

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5438] - mod_webapp doesn't work consistently

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5438

mod_webapp doesn't work consistently

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||DUPLICATE
Summary|mod_webapp doesn't work |mod_webapp doesn't work
   |consistently|consistently



--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 01:30 
---


*** This bug has been marked as a duplicate of 4334 ***

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

DO NOT REPLY [Bug 5330] - JNDI ENC context problem.

[bugzilla]

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
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=5330

JNDI ENC context problem.





--- Additional Comments From [EMAIL PROTECTED]  2001-12-17 01:30 ---
The bindings are not set for the thread which is in charge of the session
passivation, so that's why the ENC is not usable there at the moment.

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

Why Tomcat 3.3 runs much faster than 3.2.3?

[Hugh J. L.]

As I tested, Tomcat 3.3 runs much faster than 3.2.3
(for both JSP and Servlet requests). What changes are
the main factors of this performance improvement? Thanks.

__
Do You Yahoo!?
Check out Yahoo! Shopping and Yahoo! Auctions for all of
your unique holiday gifts! Buy at http://shopping.yahoo.com
or bid at http://auctions.yahoo.com

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: Configuring Multiple Tomcat JVMs with Apache - Load Balancing

[GOMEZ Henri]

>there hasn't been done anything on that topic yet ?  What's 
>the status of
>loadbalacing, either mod_jk or mod_webapp ? 
>Is that political due to if loadbalacing is working properly 
>there won't be
>any reason to take (buy) anything else than TC ?  

State of the art is that today only mod_jk could 
handle load-balancing and only when connected 
to Tomcat 3.2.x or 3.3.

Tomcat 4.0.x support ajp13 protocol, used in mod_jk
but still miss a subtil feature (jvmroute) to be
able to keep the route to the good JVM in
session mode (SessionAfinity).

But the current refactory of ajp protocol,
under ajp14 in jakarta-tomcat-connectors, will
fix somedays thanks to Costin and Kevin works :)))

--
To unsubscribe, e-mail:   
For additional commands, e-mail: 

RE: [j-t-c] patch for jakarta-tomcat-connectors/jk/native/common/jk_a jp_common.c

[GOMEZ Henri]

The patch seems valid and the pmsg was passed for this
purpose in previous release of J-T-C but some folks
(didn't remember who) make modifications in mod_jk
part of code and the update was lost.

Seems ok, you could commit :)

This patch fix a real problem and should be also 
propagated to Tomcat 3.3.1 mod_jk 

go go go

-
Henri Gomez ___[_]
EMAIL : [EMAIL PROTECTED](. .) 
PGP KEY : 697ECEDD...oOOo..(_)..oOOo...
PGP Fingerprint : 9DF8 1EA8 ED53 2F39 DC9B 904A 364F 80E6 



>-Original Message-
>From: Kevin Seguin [mailto:[EMAIL PROTECTED]]
>Sent: Thursday, December 13, 2001 11:20 PM
>To: Tomcat-Dev (E-mail)
>Subject: [j-t-c] patch for
>jakarta-tomcat-connectors/jk/native/common/jk_a jp_common.c
>
>
>this patch fixes a problem when posting with more than about 
>8K of data.  it
>seems that when the java side of the ajp connection has to ask 
>the webserver
>(jk) side for more data, it goes into an infinite loop because the next
>chunk of data isn't actually sent.
>
>i'm 97% sure about this patch, but since i don't know this 
>code as well as
>some, i'd like someone else to review it.
>
>attached is the patch and a servlet (FileUpload.java) and a client
>(Upload2.java) that demonstrate the problem.
>
>the client uses HTTPClient, which can be downloaded from
>http://www.innovation.ch/java/HTTPClient/.  incidentally, i 
>would have used
>the http client in jakarta-commons, if i could have figured 
>out how to post
>using an output stream :)
>
>this same test fails on the java side of ajp using the ajp 
>stuff on the head
>in j-t-c/jk.  again, an infinite loop.  i haven't been able to 
>figure it out
>yet...  if somebody wants extra points ... ;)
>
>thanks,
>-kevin.
>
>

--
To unsubscribe, e-mail:   
For additional commands, e-mail: