Re: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION USE???)

[Jeff Kilbride]

Hi Drasko,

Sorry, I haven't used Solaris in quite a while. I use FreeBSD and Linux most
of the time. I vaguely remember seeing posts about threading issues, but I
can't recall the problems.

I used the 3.2.x release successfully behind Apache before switching to 4.x.
I found that I didn't need Apache, except to serve static content -- no php,
perl, etc...  So, I moved to Tomcat standalone and now I run a faster,
smaller footprint webserver (thttpd) on port 81 to handle my images and
other static content. Unfortunately, you can't load balance this way without
an external solution.

Anyway, I took a look at the 3.3 ThreadPool code and I think the ThreadPool
configuration is actually being done in PoolTCPEndpoint.java in
org.apache.tomcat.util.net. This starts a pool (with the defaults) then uses
the setXXX methods to configure the Pool. When I was using 3.2.x, I remember
there being serious problems if the MAX_THREADS value was ever reached. For
heavy load, the way around it was to set MAX_THREADS equal to the number of
possible Apache child processes (MaxClients in httpd.conf). This way,
MAX_THREADS should never be reached -- hence, no blocking problem. You might
give this a try and see if it helps. It solved the problems I was having
with 3.2.x at the time.

Hope this helps.

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Sunday, December 30, 2001 9:43 PM
Subject: Re: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION USE???)


>
>
> Thanks for your reply Jeff :-)
>
> I am using it "behind" the Apache Web server (ajp1.3)  and am having
> serious problems to keep the service alive (not to mention the poor
> performance)
> The Tomcat is increasing the number of threads when under the load and
> inevitable finishes with ThreadPool giving up.
> The code in the ThreadPool is really "not okay" as it does not provide any
> way of tuning the pools and definetly does not provide any solution when
> running out of available Threads :-(
> By the way, do you have any experience with the load-balancing under
ajp1.3
> ... up to now I have observed only that one blocked Tomcat is stalling the
> whole ajp1.3 load-balancing for the rest of the gang (I have 4 Solaris
> machine each configured with identical application).
> Also browsing a bit thrue the internet I have found quite a few articles
> about Solaris and Threads in JVM ... mostly confusing me if not giving me
> very bad feeling in stomach ... any comments about the stability of the
> Tomcat3.3/JDK1.3.1_02/Solaris8 ???
>
> Thanks in advance
> Drasko
>
>
>
>
>
>
> "Jeff Kilbride" <[EMAIL PROTECTED]> on 31.12.2001 06:34:42
>
> Please respond to "Tomcat Users List" <[EMAIL PROTECTED]>
>
> To:   "Tomcat Users List" <[EMAIL PROTECTED]>
> cc:
>
> Subject:  Re: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION
USE???)
>
> Are you using Tomcat standalone or with Apache? If standalone, then yes, I
> would recommend using the latest version of Tomcat 4.x. If you're using
> Apache, I would stick with 3.3 until the mod_jk code is fully implemented
> with the 4.x versions.
>
> Thanks,
> --jeff
>
> - Original Message -
> From: <[EMAIL PROTECTED]>
> To: "Tomcat Users List" <[EMAIL PROTECTED]>
> Sent: Saturday, December 29, 2001 2:09 PM
> Subject: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION USE???)
>
>
> >
> >
> > Hi there,
> >
> > I am trying to deploy our production service on the Tomcat 3.3 platform
> > (Solaris8, JDK 1.3.1) but am very often facing the instability problems.
> > Somehow, the ThreadPool just runs out of steam and blocks Connectors
from
> > receiving any more requests.
> >
> > Looking into the source code, I have noticed some very interesting
> > "constructions" in the ThreadPool class.  Namely, the setters for the
> > maxThreads, maxSpareThreads and minSpareThreads are having very little
> > influence as the start() method is resseting the values to the default
> > constants ?!?!
> >
> > public ThreadPool() {
> > maxThreads  = MAX_THREADS;
> > maxSpareThreads = MAX_SPARE_THREADS;
> > minSpareThreads = MIN_SPARE_THREADS;
> > currentThreadCount  = 0;
> > currentThreadsBusy  = 0;
> > stopThePool = false;
> > }
> >
> > public synchronized void start() {
> >  stopThePool=false;
> > currentThreadCount  = 0;
> > currentThreadsBusy  = 0;
> > maxThreads  = MAX_THREADS;
> > maxSpareThreads = 

Re: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION USE???)

[Jeff Kilbride]

Are you using Tomcat standalone or with Apache? If standalone, then yes, I
would recommend using the latest version of Tomcat 4.x. If you're using
Apache, I would stick with 3.3 until the mod_jk code is fully implemented
with the 4.x versions.

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Saturday, December 29, 2001 2:09 PM
Subject: Tomcat 3.3: ThreadPool bug (IS IT SAFE FOR PRODUCTION USE???)


>
>
> Hi there,
>
> I am trying to deploy our production service on the Tomcat 3.3 platform
> (Solaris8, JDK 1.3.1) but am very often facing the instability problems.
> Somehow, the ThreadPool just runs out of steam and blocks Connectors from
> receiving any more requests.
>
> Looking into the source code, I have noticed some very interesting
> "constructions" in the ThreadPool class.  Namely, the setters for the
> maxThreads, maxSpareThreads and minSpareThreads are having very little
> influence as the start() method is resseting the values to the default
> constants ?!?!
>
> public ThreadPool() {
> maxThreads  = MAX_THREADS;
> maxSpareThreads = MAX_SPARE_THREADS;
> minSpareThreads = MIN_SPARE_THREADS;
> currentThreadCount  = 0;
> currentThreadsBusy  = 0;
> stopThePool = false;
> }
>
> public synchronized void start() {
>  stopThePool=false;
> currentThreadCount  = 0;
> currentThreadsBusy  = 0;
> maxThreads  = MAX_THREADS;
> maxSpareThreads = MAX_SPARE_THREADS;
> minSpareThreads = MIN_SPARE_THREADS;
>
> adjustLimits();
>
> openThreads(minSpareThreads);
> monitor = new MonitorRunnable(this);
> }
>
> Also, the situation when there are no more available Threads in the pool
is being handled in a special way which JUST BLOCKS IN THE INFINITE LOOP :
> -(((
>
> Is the code of the Tomcat4 equaly immature or there is a fair chance to
> role a productiv system on top of it?
>
> Thanks in advance for any comments
>
> Drasko Kokic
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: How to Increasing Tomcat memory?

[Jeff Kilbride]

Hi Jack,

You should use the TOMCAT_OPTS environment variable to set these options.
So, create an environment variable called TOMCAT_OPTS and set it equal to
the following:

"-Xms32M -Xmx64M"

This will set the initial size of the JVM heap to 32MB and the max size to
64MB. Replace these values with whatever you require and you're all set. The
startup script for Tomcat looks for the TOMCAT_OPTS env. variable and
automatically appends it's contents to the command line when starting the
JVM.

I recommend that you upgrade to at least version 3.2.4, due to security bugs
that have been fixed in 3.2.1 and 3.2.2. None of your configuration files
need to be changed. Just swap out the jar files in your TOMCAT_HOME/lib
directory with the new ones.

Thanks,
--jeff

- Original Message -
From: "Jack Li" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 1:44 PM
Subject: How to Increasing Tomcat memory?


> Hi,
>
> I need to allocate more memory to Tomcat 3.2.1. How can I do it? I know to
> put -Xms somewhere. What is the exact lines to put and where to put the
> line?
>
> Thanks,
> Jack
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: There's 6 mod_jk's? Which one?

[Jeff Kilbride]

You can also run your httpd binary from the command line with the -V switch:

[root@www /root]# httpd -V

This will print out all the options that your apache binary was compiled
with. If you see a line like this:

 -D EAPI

Then your apache binary was compiled with EAPI and you need the EAPI version
of mod_jk. If not, use the noeapi version.

Thanks,
--jeff

- Original Message -
From: "Allan Tomalesky" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, December 17, 2001 1:41 PM
Subject: Re: There's 6 mod_jk's? Which one?


> Hi,
>
> If you are using  apache 1.3.x use ap13 (apache 2.0 try ap20).  Try each
> of the two  *.so, if it's the wrong one you will get an error message
> (eapi or noeapi) depending on which apache binary you are using.
>
> Good luck.
> -allan
>
> Scott Merritt wrote:
>
> >Ok...  I see in the compiled dir for Linux, 6 mod_jk's...  How do I
> >determine if my Apache web server is compiled with eapi or not?  Also how
do
> >I know if I get the one for ap13 or ap20?
> >
> >I'm using Tomcat 4...
> >
> > mod_jk-3.3-ap13-eapi.so23-Oct-2001 12:52   93K
> > mod_jk-3.3-ap13-eapi.so.asc23-Oct-2001 12:53  285
> > mod_jk-3.3-ap13-noeapi.so  23-Oct-2001 12:53   93K
> > mod_jk-3.3-ap13-noeapi.so.asc  23-Oct-2001 12:53  285
> > mod_jk-3.3-ap20.so 23-Oct-2001 12:54  114K
> > mod_jk-3.3-ap20.so.asc
> >
> >Thanks..
> >
> >--
> >To unsubscribe:   
> >For additional commands: 
> >Troubles with the list: 
> >
> >
> >
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Ever increasing heap size with Tomcat 3.2.3 !!!

[Jeff Kilbride]

I believe you need to supply the '-server' switch at startup to the JVM to
be in "server mode". Try dropping that switch, if you're using it.

Personally, I recommend you check out the IBM JDK, too. You can install both
and easily switch between them by modifying your JAVA_HOME environment
variable.

Thanks,
--jeff

- Original Message -
From: "Hawkins, Keith (Keith)" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 10:09 AM
Subject: RE: Ever increasing heap size with Tomcat 3.2.3 !!!


> John,
> Thanks for the reply.  How do I tell whether I am using the server JVM
> or
> the client JVM you mentioned?
> Thanks,
> Keith
>
> -Original Message-
> From: John Freeborg [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, December 18, 2001 1:08 PM
> To: Tomcat Users List
> Subject: RE: Ever increasing heap size with Tomcat 3.2.3 !!!
>
>
> Which JVM are you using?
>
> On Windows 2K w/ SP2 I found that the Sun JDK 1.3.1 server hotspot JVM
> crashed and burned running Tomcat 4.0.1 this way within 24 hours easily.
> A few others emailed me about it also.
>
> Switching to the Sun JDK 1.3.1 client hotspot JVM magically fixed this.
> Now my server runs for weeks without crashing.
>
> Might be the same issue with Tomcat 3.2.3 - try it.  I was going to try
> an IBM JDK also, but never got around to it once I had it working.
>
>  - John
>
> -Original Message-
> From: Randy Layman [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, December 18, 2001 11:07 AM
> To: Tomcat Users List
> Subject: RE: Ever increasing heap size with Tomcat 3.2.3 !!!
>
>
>
>
> > -Original Message-
> > From: Hawkins, Keith (Keith) [mailto:[EMAIL PROTECTED]]
> > Sent: Tuesday, December 18, 2001 12:34 PM
> > To: [EMAIL PROTECTED]; tomcat-user
> > Subject: Ever increasing heap size with Tomcat 3.2.3 !!!
> >
> >
> >
> > Are there known issues with Tomcat and heap size??
>
> No.
> >
> > Doing a web search revealed numerous posts with people having similar
> > problems so I believe there is a problem.   The standard
> > response these
> > people receive is to increase the heap size via -Xmx   But that seems
> > like a band-aid rather than a real solution.   That just delays the
> > inevitable.
>
> The problem is always that they are holding onto memory that they don't
> realize they are holding on to, or are expecting to be garbage collected
> but
> can't for whatever reason.
>
> Here are some suggestions on where to look:
> 1.  Don't use class variables in servlets or JSPs
> 2.  Be careful with sessions.  Setting the inactive timeout to
> nothing allows the sessions to stay around until the server is reset,
> and
> sessions last for some time after the last request
> 3.  Understand that it anything has a reference to an object
> (list,
> map, array) then it can't be garbage collected
> 4.  Use a program like OptimizeIT! to find your memory leak
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Stability of Tomcat 3.3 vs. 4.0

[Jeff Kilbride]

Unfortunately, Tomcat 4.x doesn't support load-balancing, yet -- even with
mod_jk. So, if you need it, you should stick with 3.3.

Thanks,
--jeff

- Original Message -
From: "Hunter Hillegas" <[EMAIL PROTECTED]>
To: "Tomcat User List" <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 10:31 AM
Subject: Re: Stability of Tomcat 3.3 vs. 4.0


> I would suggest Tomcat 4.0 because it is integrated into Jboss which gives
> you a MAJOR speed up between the servlet layer and the Jboss layer. Major
> speed up.
>
> > From: "Wilson, Wayne" <[EMAIL PROTECTED]>
> > Reply-To: "Tomcat Users List" <[EMAIL PROTECTED]>
> > Date: Tue, 18 Dec 2001 13:21:34 -0500
> > To: "Tomcat Users List (E-mail)" <[EMAIL PROTECTED]>
> > Subject: Stability of Tomcat 3.3 vs. 4.0
> >
> > Hi,
> >
> > Because of some deadlock bugs in Tomcat 3.2.3 that we are experiencing
(891,
> > 1006 & 1798), we are going to have to switch to either Tomcat 3.3 or
4.X.
> > What do people think about the stability of each of these versions? We
will
> > be running this on W2K in conjunction with JBoss and because we need
load
> > balancing we will be using mod_jk.
> >
> > What do you think?
> >
> > --Wayne
> >
> > --
> > To unsubscribe:   
> > For additional commands: 
> > Troubles with the list: 
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: This is unbelievably Hard, please help!!!

[Jeff Kilbride]

Hi Brandon,

I recommend that you use the mod_jk from the 3.3 distribution. I'm not sure
if all the bug fixes have been back-ported to the 3.2.x version. Also, the
3.3 version lets you restart Tomcat without having to restart Apache.

Anyway, download the 3.3 bundle (jakarta-tomcat-3.3-src.tar.gz) from:

http://jakarta.apache.org/builds/jakarta-tomcat/release/v3.3/src/

gunzip and untar it and then cd to the src/native/mod_jk/apache1.3
directory. Copy the "Makefile.linux" to "Makefile" and then type "make".
Took about 10 seconds to compile on my stock RedHat 7.1 installation. Copy
the resulting "mod_jk.so" to your apache libexec directory and you should be
set.

Thanks,
--jeff

- Original Message -
From: "Brandon Cruz" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Wednesday, December 12, 2001 9:46 AM
Subject: This is unbelievably Hard, please help!!!


>
> I am trying to compile mod_jk.c on linux redhat 7.1.  I seem to have
gotten
> the file to compile after making some changes in mod_jk.c, the resulting
> mod_jk.so is 18000 bytes.  Apache will not start when using this file.  I
am
> using the tomcat 3.2.4 documentation to compile.  I see that there is a
> Makefile.linux in the directory, but this does not work either.  Is there
> anyone that can help me to accomplish this task?  I am stuck and have been
> working on it forever!!!
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Change an include file requires re-compiling all JSP that include it (?)

[Jeff Kilbride]

If your included files don't contain any JSP code, you can use the
 directive. This directive includes the file at request
time, rather than page translation time -- so changes to your included file
will show up immediately.

If your included files contain JSP code, then you're stuck with using the
<%@ include... %> and you'll need to recompile all your JSPs whenever the
included file changes.

Thanks,
--jeff

- Original Message -
From: "Scott Hodson" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Tuesday, December 11, 2001 10:58 AM
Subject: RE: Change an include file requires re-compiling all JSP that
include it (?)


> 1) Where do they go?  I don't see them anywhere under my webapp's folder
(I
> thought they go in WEB-INF/classes but they're not there)
>
> 2) Blech, that's what I do now.  If I have 100 JSP files all including the
> same header file I'm doomed!
>
> -Original Message-
> From: James Chuang [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, December 11, 2001 10:06 AM
> To: Tomcat Users List
> Subject: Re: Change an include file requires re-compiling all JSP that
> include it (?)
>
>
> A couple of things you can do, both would be easier than resaving all the
> JSPs
>
> 1. Delete the generated class files...
>
> 2. Touch the JSP files.
>
>
> - Original Message -
> From: "Scott Hodson" <[EMAIL PROTECTED]>
> To: "Tomcat User" <[EMAIL PROTECTED]>
> Sent: Tuesday, December 11, 2001 10:00 AM
> Subject: Change an include file requires re-compiling all JSP that include
> it (?)
>
>
> > I come from an ASP background where we would frequently change include
> files
> > and see the results immediately.  However, in JSP, since JSP pages are
> > compiled into servlet classes, if I change an include file the JSP file
> > including it won't get re-compiled because the JSP hasn't changed, just
> the
> > include file.  Even if I restart Tomcat it still won't recompile the
JSPs.
> > So for now every time I make a change to an include file I have to
re-save
> > all of my JSPs so Tomcat forces a recompile.  That's a big pain.
> >
> > Can somebody help me out here?  Is there a way to force re-compilation
of
> > JSPs if the files they include ever change?
> > ___
> >
> > Scott Hodson
> > (949) 709-4496 office
> > (949) 709-3890 fax
> > [EMAIL PROTECTED]
> > http://www.ubero.com
> >
> >
> >
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

We're going an awfully long way here to solve what could be cleanly done
with a simple:



Don't you think? All the container has to do is default the MIME type for
unknown extensions to "text/html". Remember this is a JSP -- a scripting
language which was originally designed to simplify life for web designers
and other non-programmers. Having to open every included file manually with
java code defeats that purpose.

Thanks,
--jeff

- Original Message -
From: "August Detlefsen" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, December 10, 2001 1:21 PM
Subject: RE: Unable to include *.sum files (Again)


> What text format are your .sum files using? Sounds like it may be UTF8.
> Try this:
>
>/**
> * Returns the String contents of a UTF8 file.
> *
> * This method throws any fileIO errors.
> *
> * @param sFileName   Full file path.
> * @return String   The contents of the file as a String object.
> * @throws Exception   Any fileIO errors
> */
> public static String getUTF8FileAsString(String sFileName) throws
> Exception {
> RandomAccessFile inputFile = new
> RandomAccessFile(sFileName,"r");
> String output = inputFile.readUTF();
> inputFile.close();
> return output;
> }
>
>
> This method should work if your files are in ASCII:
>
>/**
> * Returns the String contents of an ASCII file.
> *
> * This method throws any fileIO errors.
> *
> * @param sFileName   Full file path.
> * @return String   The contents of the file as a String object.
> * @throws Exception   Any fileIO errors
> */
> public static String getFileAsString(String sFileName) throws
> Exception {
> RandomAccessFile inputFile = new
> RandomAccessFile(sFileName,"r");
> byte[] inputbytes = new byte[(int)inputFile.length()];
> int numread = inputFile.read(inputbytes);
> inputFile.close();
> return new String(inputbytes);
> }
>
>
>
> --- Micael Padraig Og mac Grene <[EMAIL PROTECTED]> wrote:
> > At 08:14 AM 12/10/01 +0200, you wrote:
> > >Hi all,
> > >
> > >Thanks for all the responses(including the debate) to this question!
> > It made
> > >some real interesting reading material after the SHORT weekend! To
> > get back
> > >to August's suggestion: we've tried it but our problem is that the
> > file
> > >content is generated by a VB program and
> > >contains some funny characters e.g. "CPI" rather than "CPI". When we
> > >translate these to a string it either comes out as ?CPI? or as
> > illustrated
> > >in the attached image(This is also how
> > >it displays in JBuilder).
> > >
> > >Regarding the debate I tend to agree with Jeff. If you want to
> > display the
> > >pure contents of a file you should be able to include the file using
> > > without having to define a mime type. I mean what
> > happens if
> > >you want to include a "code" example, for example a code snippet
> > that
> > >illustrates how to code something in C,C++,Java etc. If you define
> > the mime
> > >type it will try to translate it, which is not what we want in this
> > case...
> > >You could define it as type text but now you need to maintain two
> > mime types
> > >for one extension? Just doesn't sound right to me. The other thing
> > that
> > >bothers me is the fact that it works for the <%@ include...%>
> > directive but
> > >not for the  surely they should perform similar
> > actions
> > >simply using a different syntax?
> > >
> > >Thanks again,
> > >Jonathan
> >
> > I still don't see, Jonathan, why you don't just use code in your
> > include
> > which catches the mime types and deals with them?  Why is the include
> >
> > important to you in the first instance?  I think the people in this
> > list
> > might be able to help you, if we knew what the facets of the problem
> > are.  This sounds like a problem that can be solved, but I am not
> > sure what
> > the situation is.
> >
> > Micael
> >
> >
> > --
> > To unsubscribe:   
> > For additional commands: 
> > Troubles with the list: 
> >
>
>
> __
> Do You Yahoo!?
> Send your FREE holiday greetings online!
> http://greetings.yahoo.com
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Hi Jonathon,

Well, obviously, I agree with you. I think the most interesting point you
bring up is the fact that <%@ include...%> works with any file extension,
whether or not it has a registered MIME type. Why should 
add additional restrictions? I understand the difference between the two --
page translation time vs. request time -- but don't understand why they act
differently. I've gone through the JSP spec for both of them and can find
nothing that requires the MIME type of the included file to be a registered
text/* type.

Can anybody give a plausible reason why one requires the MIME type to be
registered while the other does not?

Thanks,
--jeff

- Original Message -
From: "Kusel, Jonathan J" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Sunday, December 09, 2001 10:14 PM
Subject: RE: Unable to include *.sum files (Again)


> Hi all,
>
> Thanks for all the responses(including the debate) to this question! It
made
> some real interesting reading material after the SHORT weekend! To get
back
> to August's suggestion: we've tried it but our problem is that the file
> content is generated by a VB program and
> contains some funny characters e.g. "CPI" rather than "CPI". When we
> translate these to a string it either comes out as ?CPI? or as illustrated
> in the attached image(This is also how
> it displays in JBuilder).
>
> Regarding the debate I tend to agree with Jeff. If you want to display the
> pure contents of a file you should be able to include the file using
>  without having to define a mime type. I mean what happens if
> you want to include a "code" example, for example a code snippet that
> illustrates how to code something in C,C++,Java etc. If you define the
mime
> type it will try to translate it, which is not what we want in this
case...
> You could define it as type text but now you need to maintain two mime
types
> for one extension? Just doesn't sound right to me. The other thing that
> bothers me is the fact that it works for the <%@ include...%> directive
but
> not for the  surely they should perform similar actions
> simply using a different syntax?
>
> Thanks again,
> Jonathan
>
> -Original Message-
> From: August Detlefsen [mailto:[EMAIL PROTECTED]]
> Sent: 09 December 2001 01:53
> To: Tomcat Users List
> Subject: Re: Unable to include *.sum files (Again)
>
>
> If it does use PrintWriter to write the output, then it makes sense for
> it to only output text. See this from the PrintWriter javadoc:
>
> Print formatted representations of objects to a text-output stream.
> This class implements all of the print methods found in PrintStream. It
> does not contain methods for writing raw bytes, for which a program
> should use unencoded byte streams.
>
> If you want to include files of non-text types (or types that are text,
> but not included in your MIME types list), why not just write a utility
> method that opens a file, reads it and returns its contents as a
> String?
>
> -August
>
> --- Micael Padraig Og mac Grene <[EMAIL PROTECTED]> wrote:
> > At 08:22 PM 12/7/01 -0800, you wrote:
> > >Yeah, see my last post. Since JSP output is written with a
> > PrintWriter, the
> > >Catalina code is restricting it to only being able to output known
> > text/*
> > >MIME types. This just doesn't "feel" right to me.
> > >
> > >Thanks,
> > >--jeff
> >
> > Well, jeff, then it is not a "bug".  At best it is a difference of
> > opinion.  That makes all the sense in the world to me.  If you want
> > to
> > bring in something other than the "known 'text/* MIME types," just
> > include
> > the proper code in your include?  We have differing intuitions here.
> > I
> > think what Catalina is doing is proper and makes sense.  But, at
> > worst for
> > you, it is an inconvenience.  Right?
> >
> > -- micael
> >
> >
> > --
> > To unsubscribe:   
> > For additional commands: 
> > Troubles with the list: 
> >
>
>
> __
> Do You Yahoo!?
> Send your FREE holiday greetings online!
> http://greetings.yahoo.com
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>






>
>
>
> Disclaimer and Confidentiality Note.
>
> Everything in this e-mail and attachments relating to the
> official business of Standard Bank Investment Corporation(Stanbic)
> is proprietary to the company. It is confidential, legally privileged
> and protected by law. Stanbic does not own and endorse
> any other content. Views and opinions are those of the
> sender unless clearly stated as being that of Stanbic.
>
> The person addressed in the e-mail is the sole authorised recipient.
> Please notify the sender immediately if it has unintentionally reached
> you and do not read, disclose or use

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

I agree that there are workarounds -- there are always workarounds -- but
Tomcat is the RI, so this is the place where we're supposed to figure this
stuff out.

I guess my argument is that Tomcat is being more restrictive than the spec
requires it to be. Whether that's acceptable or not is ultimately up to the
tomcat dev team.

Thanks,
--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Saturday, December 08, 2001 4:27 PM
Subject: Re: Unable to include *.sum files (Again)


> At 03:53 PM 12/8/01 -0800, you wrote:
> >If it does use PrintWriter to write the output, then it makes sense for
> >it to only output text. See this from the PrintWriter javadoc:
> >
> >Print formatted representations of objects to a text-output stream.
> >This class implements all of the print methods found in PrintStream. It
> >does not contain methods for writing raw bytes, for which a program
> >should use unencoded byte streams.
> >
> >If you want to include files of non-text types (or types that are text,
> >but not included in your MIME types list), why not just write a utility
> >method that opens a file, reads it and returns its contents as a
> >String?
> >
> >-August
>
> Yah, August, we all agree on that.  Jeff is thinking that the
> specifications are not clear enough.  Just a difference of "style" I
> suppose.  Thanks for your input.
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Ok, but what happens if you don't have control of the included file? Your
database or other source produces dynamic reports for you and you need to
include them into your JSPs.

My question is this: if nearly every webserver on the planet can use a
default MIME type of text/html, why can't this? That would solve the
problem. Restricting output to registered MIME types seems a little heavy
handed to me.

Thanks,
--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Saturday, December 08, 2001 2:58 PM
Subject: Re: Unable to include *.sum files (Again)


> At 02:51 PM 12/8/01 -0800, you wrote:
> >I don't think the spec is that detailed -- I mean, it doesn't come out
and
> >say the "page" attribute of jsp:include has to follow the requirements of
> >the JspWriter. So, I don't know the answer to that.
> >
> >Remember though, we're talking about included files -- by their very
nature,
> >they don't necessarily represent *entire* files. They may only be pieces
of
> >a bigger file that is ultimately displayed to the surfer. I can
understand
> >that the file sent to the browser must have a sensible MIME type, but
must
> >all the pieces that the file is built from also have registered MIME
types?
> >I'm not sure that makes sense.
> >
> >Also, I always thought MIME types, in this context, were available so the
> >browser could decipher what kind of file was being retrieved and display
> >that file correctly. I never thought MIME types would be used to limit my
> >flexibility with JSPs in this way.
> >
> >Thanks,
> >--jeff
>
> I understand what you are saying, Jeff.  But the bottom line is that the
> JspWriter is the source of the out implicit object in this
> instance.  Apparently your bottom line is that you want to use the
> extension to pass information, instead of an extension in any true
> sense.  I think it is a little harsh to have them "expect" that use.  That
> said, and I may be wrong, why not toss the '"dynamic" values in prior to
> the extension and trick the JspWriter, e.g. instead of com.site.123 us
> com.site.123.txt or com.site.123.jsp.  You are going to have to have a
> class read the url in any event.  The other way is to put your dynamic
> information inside the included tidbit.  That seems more intuitive to me
> than using an extension for information anyway.  Just a thought.
>
> -- micael
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

I don't think the spec is that detailed -- I mean, it doesn't come out and
say the "page" attribute of jsp:include has to follow the requirements of
the JspWriter. So, I don't know the answer to that.

Remember though, we're talking about included files -- by their very nature,
they don't necessarily represent *entire* files. They may only be pieces of
a bigger file that is ultimately displayed to the surfer. I can understand
that the file sent to the browser must have a sensible MIME type, but must
all the pieces that the file is built from also have registered MIME types?
I'm not sure that makes sense.

Also, I always thought MIME types, in this context, were available so the
browser could decipher what kind of file was being retrieved and display
that file correctly. I never thought MIME types would be used to limit my
flexibility with JSPs in this way.

Thanks,
--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Saturday, December 08, 2001 2:06 PM
Subject: Re: Unable to include *.sum files (Again)


> At 01:27 PM 12/8/01 -0800, you wrote:
> >I'm not so sure. The JSP spec doesn't say anything about having to
register
> >the MIME type of a file before using it in a jsp:include directive.
Consider
> >how difficult that would be if you were generating dynamic filenames with
> >extensions like ".001", ".002", ".003", etc... You could potentially have
to
> >register several hundred (even thousand) meaningless MIME types. Also,
think
> >about what webservers do when they encounter an unknown MIME type -- they
> >default back to text/html (or whatever you have set as your default...).
Why
> >should a JSP be more strict than this with it's include directive? The
file
> >you are including is required to be a valid URL -- which means it could
be
> >accessed via the webserver anyway.
> >
> >Remember, Tomcat is the RI and should implement the spec as closely as
> >possible. I've gone back and looked through my books on JSP, and every
one
> >of them says that you can use any file extension with the jsp:include
> >directive. However, none of them say you have to register that extension
as
> >a valid MIME type first. At the very least, it's confusing for someone
who
> >expects it to work when it doesn't -- especially when there's no
> >documentation on it.
> >
> >Thanks,
> >--jeff
> >
> >- Original Message -
> >From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
> >To: "Tomcat Users List" <[EMAIL PROTECTED]>
> >Sent: Saturday, December 08, 2001 10:12 AM
> >Subject: Re: Unable to include *.sum files (Again)
> >
> >
> > > At 08:22 PM 12/7/01 -0800, you wrote:
> > > >Yeah, see my last post. Since JSP output is written with a
PrintWriter,
> >the
> > > >Catalina code is restricting it to only being able to output known
text/*
> > > >MIME types. This just doesn't "feel" right to me.
> > > >
> > > >Thanks,
> > > >--jeff
> > >
>
> The JSP specification does tie the use of include to the requirements of
> the out JspWriter ?
> --- micael
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

I'm not so sure. The JSP spec doesn't say anything about having to register
the MIME type of a file before using it in a jsp:include directive. Consider
how difficult that would be if you were generating dynamic filenames with
extensions like ".001", ".002", ".003", etc... You could potentially have to
register several hundred (even thousand) meaningless MIME types. Also, think
about what webservers do when they encounter an unknown MIME type -- they
default back to text/html (or whatever you have set as your default...). Why
should a JSP be more strict than this with it's include directive? The file
you are including is required to be a valid URL -- which means it could be
accessed via the webserver anyway.

Remember, Tomcat is the RI and should implement the spec as closely as
possible. I've gone back and looked through my books on JSP, and every one
of them says that you can use any file extension with the jsp:include
directive. However, none of them say you have to register that extension as
a valid MIME type first. At the very least, it's confusing for someone who
expects it to work when it doesn't -- especially when there's no
documentation on it.

Thanks,
--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Saturday, December 08, 2001 10:12 AM
Subject: Re: Unable to include *.sum files (Again)


> At 08:22 PM 12/7/01 -0800, you wrote:
> >Yeah, see my last post. Since JSP output is written with a PrintWriter,
the
> >Catalina code is restricting it to only being able to output known text/*
> >MIME types. This just doesn't "feel" right to me.
> >
> >Thanks,
> >--jeff
>
> Well, jeff, then it is not a "bug".  At best it is a difference of
> opinion.  That makes all the sense in the world to me.  If you want to
> bring in something other than the "known 'text/* MIME types," just include
> the proper code in your include?  We have differing intuitions here.  I
> think what Catalina is doing is proper and makes sense.  But, at worst for
> you, it is an inconvenience.  Right?
>
> -- micael
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Yeah, see my last post. Since JSP output is written with a PrintWriter, the
Catalina code is restricting it to only being able to output known text/*
MIME types. This just doesn't "feel" right to me.

Thanks,
--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 7:40 PM
Subject: Re: Unable to include *.sum files (Again)


>
>
> At 02:21 PM 12/7/01 -0800, you wrote:
> >Hi Micael,
> >
> >Let's make sure we're talking about the same thing. If I include a file
> >like:
> >
> >
> >
> >Why does it matter what the file extension of the included file is? Isn't
> >the container just supposed to open the file and output it's contents at
> >that particular spot in my JSP? Now, I agree that the contents of the
> >included file will be affected by the MIME type of the response -- how
the
> >browser views those contents. If you include a file that doesn't match
the
> >MIME type of your response, it may not appear correctly in your browser.
But
> >that's not the point.
> >
> >The JSP spec says the "page" attribute must evaluate to a String that is
a
> >relative URL specification. When I pull up the testTxt.sum file in my
> >browser, it displays correctly (it's only one line of text...). So, I
should
> >be able to include it in my JSP with the jsp:include directive. When I
> >change the extension to .txt, .html, or .jsp it works. Anything else,
even
> >other registered MIME types like .doc, and it doesn't. (ok, I didn't try
> >them ALL...  :)
> >
> >I think this is a bug. The old Java Web Server used to have a similar bug
> >where it would only include .htm and .html files.
> >
> >--jeff
>
> Before I look further, have you looked at the source code to see why this
> is happening?
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Does Tomcat support Server Push?

[Jeff Kilbride]

It's not Tomcat, it's the browser. IE doesn't support server push -- which I
think is stupid, but I'm sure there's a reason. Netscape is the only browser
I know that supports multipart/x-mixed-replace.

--jeff

- Original Message -
From: "Bill Taylor" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 10:13 AM
Subject: Does Tomcat support Server Push?


> A content type of
>
> multipart/x-mixed-replace;boundary=End\n
> \n
> --End\n
> Content type: text/html\n
> 
> --End\n
> \n
>
> Supposedly lets the server push a new page image to the client at will.
>
> I have tried to do this with Tomcat by setting the content type and
> sending successive pages in a loop, but all the pages come out as text
> at the same time.
>
> Is there a known way to support server push with Tomcat?
>
> Thanks.
>
> Bill Taylor
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

If it does, then it's violating the spec. The spec doesn't specify any
extensions that can or cannot be included. It just says it has to be a
String that is a relative URL. The string "./myText.foo" satisfies that and
the file myText.foo pulls up correctly in my browser with the following
content:

"This is the myText.foo file!"

However, when you include the myText.foo file using the jsp:include
directive, you get an exception. I think that's a bug.

My setup is TC 4.0.1 stand-alone on RedHat 7.1 using the IBM JDK 1.2 -- in
case anyone was wondering...

--jeff

- Original Message -
From: "August Detlefsen" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 2:16 PM
Subject: Re: Unable to include *.sum files (Again)


> It might, if somewhere in the code of the  tag it
> specifies to only include a file if it is of a relevant mime type, say
> JSP, text/xyz, or image/abc.
>
> This would prevent the server from sending out potentially damaging
> stuff with the include tag, like .java or .exe files.
>
> Again, I am just speculating...
>
>
>
>
> --- Jeff Kilbride <[EMAIL PROTECTED]> wrote:
> > Nope. I tried it with *.doc files, too, and it still doesn't work.
> > *.doc is
> > defined in web.xml.
> >
> > Besides, it doesn't really make sense for MIME types to affect
> > included
> > files, does it?
> >
> > --jeff
> >
> > - Original Message -
> > From: "August Detlefsen" <[EMAIL PROTECTED]>
> > To: "Tomcat Users List" <[EMAIL PROTECTED]>
> > Sent: Friday, December 07, 2001 12:40 PM
> > Subject: Re: Unable to include *.sum files (Again)
> >
> >
> > > I am totally guessing, but maybe you have to define a MIME type for
> > > .sum files in your web.xml (and perhaps also where Apache
> > configures
> > > MIME types?)?
> > >
> > > HTH,
> > > -August
> > >
> > > --- "Kusel, Jonathan J" <[EMAIL PROTECTED]> wrote:
> > > >
> > > > Hi all,
> > > > I've mailed the follwoing to the tomcat users group but nobody
> > has
> > > > responded
> > > > yet so I took the liberty of posting it again.
> > > >
> > > > Here's the message:
> > > > >
> > > >
> > > > I'm trying to include normal text files with ".sum"  extensions
> > using
> > > > the
> > > > jsp:include directive:
> > > >''
> > > > but tomcat throws an exception(see bottom for the exception).
> > When I
> > > > change
> > > > the file's extension to .txt for example it works, but the files
> > that
> > > > I
> > > > wan't to include are generated by another program on a daily
> > basis so
> > > > changing the extension is not an option.
> > > >
> > > > When I use the '<%@include file="testTxt.sum" %>' directive it
> > works,
> > > > but I
> > > > need to use the  since the filenames are dynamically
> > > > generated.
> > > >
> > > > Is this a bug or am I doing something wrong? I'm using tomcat4.0
> > for
> > > > NT 4.0.
> > > >
> > > > THE STACKTRACE:
> > > > 
> > > >
> > > > ApplicationDispatcher[/interestrates]: Servlet.service() for
> > servlet
> > > > default
> > > > threw exception
> > > > java.lang.IllegalStateException
> > > > at
> > > >
> > >
> >
>
org.apache.jasper.runtime.ServletResponseWrapperInclude.getOutputStream(Serv
> > > > letResponseWrapperInclude.java:109)
> > > > at
> > > >
> > >
> >
>
org.apache.catalina.servlets.DefaultServlet.serveResource(DefaultServlet.jav
> > > > a:1143)
> > > > at
> > > >
> > >
> >
> org.apache.catalina.servlets.DefaultServlet.doGet(DefaultServlet.java:519)
> > > > at javax.servlet.http.HttpServlet.service(HttpServlet.java:740)
> > > > at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> > > > at
> > > >
> > >
> >
>
org.apache.catalina.core.ApplicationDispatcher.invoke(ApplicationDispatcher.
> > > > java:679)
> > > > at
> > > >
> > >
> >
>
org.apache.c

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Hi Micael,

Let's make sure we're talking about the same thing. If I include a file
like:



Why does it matter what the file extension of the included file is? Isn't
the container just supposed to open the file and output it's contents at
that particular spot in my JSP? Now, I agree that the contents of the
included file will be affected by the MIME type of the response -- how the
browser views those contents. If you include a file that doesn't match the
MIME type of your response, it may not appear correctly in your browser. But
that's not the point.

The JSP spec says the "page" attribute must evaluate to a String that is a
relative URL specification. When I pull up the testTxt.sum file in my
browser, it displays correctly (it's only one line of text...). So, I should
be able to include it in my JSP with the jsp:include directive. When I
change the extension to .txt, .html, or .jsp it works. Anything else, even
other registered MIME types like .doc, and it doesn't. (ok, I didn't try
them ALL...  :)

I think this is a bug. The old Java Web Server used to have a similar bug
where it would only include .htm and .html files.

--jeff

- Original Message -
From: "Micael Padraig Og mac Grene" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 1:37 PM
Subject: Re: Unable to include *.sum files (Again)


> At 01:31 PM 12/7/01 -0800, you wrote:
> >Nope. I tried it with *.doc files, too, and it still doesn't work. *.doc
is
> >defined in web.xml.
> >
> >Besides, it doesn't really make sense for MIME types to affect included
> >files, does it?
> >
> >--jeff
>
> I may be out to lunch here, Jeff, but it seems to me that it makes all the
> sense in the world for an included file to be affected by its MIME type.
>
> -- micael
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Nope. I tried it with *.doc files, too, and it still doesn't work. *.doc is
defined in web.xml.

Besides, it doesn't really make sense for MIME types to affect included
files, does it?

--jeff

- Original Message -
From: "August Detlefsen" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 12:40 PM
Subject: Re: Unable to include *.sum files (Again)


> I am totally guessing, but maybe you have to define a MIME type for
> .sum files in your web.xml (and perhaps also where Apache configures
> MIME types?)?
>
> HTH,
> -August
>
> --- "Kusel, Jonathan J" <[EMAIL PROTECTED]> wrote:
> >
> > Hi all,
> > I've mailed the follwoing to the tomcat users group but nobody has
> > responded
> > yet so I took the liberty of posting it again.
> >
> > Here's the message:
> > >
> >
> > I'm trying to include normal text files with ".sum"  extensions using
> > the
> > jsp:include directive:
> >''
> > but tomcat throws an exception(see bottom for the exception). When I
> > change
> > the file's extension to .txt for example it works, but the files that
> > I
> > wan't to include are generated by another program on a daily basis so
> > changing the extension is not an option.
> >
> > When I use the '<%@include file="testTxt.sum" %>' directive it works,
> > but I
> > need to use the  since the filenames are dynamically
> > generated.
> >
> > Is this a bug or am I doing something wrong? I'm using tomcat4.0 for
> > NT 4.0.
> >
> > THE STACKTRACE:
> > 
> >
> > ApplicationDispatcher[/interestrates]: Servlet.service() for servlet
> > default
> > threw exception
> > java.lang.IllegalStateException
> > at
> >
>
org.apache.jasper.runtime.ServletResponseWrapperInclude.getOutputStream(Serv
> > letResponseWrapperInclude.java:109)
> > at
> >
>
org.apache.catalina.servlets.DefaultServlet.serveResource(DefaultServlet.jav
> > a:1143)
> > at
> >
> org.apache.catalina.servlets.DefaultServlet.doGet(DefaultServlet.java:519)
> > at javax.servlet.http.HttpServlet.service(HttpServlet.java:740)
> > at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> > at
> >
>
org.apache.catalina.core.ApplicationDispatcher.invoke(ApplicationDispatcher.
> > java:679)
> > at
> >
>
org.apache.catalina.core.ApplicationDispatcher.doInclude(ApplicationDispatch
> > er.java:570)
> > at
> >
>
org.apache.catalina.core.ApplicationDispatcher.include(ApplicationDispatcher
> > .java:493)
> > at
> >
>
org.apache.jasper.runtime.JspRuntimeLibrary.include(JspRuntimeLibrary.java:8
> > 18)
> > at org.apache.jsp.test$jsp._jspService(test$jsp.java:67)
> > at
> > org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:107)
> > at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> > at
> >
>
org.apache.jasper.servlet.JspServlet$JspServletWrapper.service(JspServlet.ja
> > va:202)
> > at
> >
> org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:382)
> > at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:474)
> > at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> > at
> >
>
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(Application
> > FilterChain.java:247)
> > at
> >
>
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterCh
> > ain.java:193)
> > at
> >
>
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.ja
> > va:243)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> > 66)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> > at
> > org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> > at
> >
>
org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.ja
> > va:201)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> > 66)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> > at
> > org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> > at
> >
> org.apache.catalina.core.StandardContext.invoke(StandardContext.java:2344)
> > at
> >
>
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:164
> > )
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> > 66)
> > at
> >
>
org.apache.catalina.valves.ErrorDispatcherValve.invoke(ErrorDispatcherValve.
> > java:170)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> > 64)
> > at
> >
>
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:170
> > )
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> > 64)
> > at
> >
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> > at
> > org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> > at
> >
>
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngine

Re: Unable to include *.sum files (Again)

[Jeff Kilbride]

Just for fun, I tested this with the snoop.jsp file included in the examples
by adding:



I then created a one-line testTxt.sum file in the same directory with the
following contents:

Successfully included testTxt.sum!

You're right! It doesn't work. I get the same exception in my logs. I then
tried several different extensions and found that the only ones that seem to
work are .jsp, .html, and .txt.

This is a bug.

For the time being, can you copy your *.sum files to *.txt in your JSP prior
to the jsp:include directive. It's a hack, but it should work until this is
resolved.

--jeff

- Original Message -
From: "Kusel, Jonathan J" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 2:22 AM
Subject: Unable to include *.sum files (Again)


>
> Hi all,
> I've mailed the follwoing to the tomcat users group but nobody has
responded
> yet so I took the liberty of posting it again.
>
> Here's the message:
> >
>
> I'm trying to include normal text files with ".sum"  extensions using the
> jsp:include directive:
>''
> but tomcat throws an exception(see bottom for the exception). When I
change
> the file's extension to .txt for example it works, but the files that I
> wan't to include are generated by another program on a daily basis so
> changing the extension is not an option.
>
> When I use the '<%@include file="testTxt.sum" %>' directive it works, but
I
> need to use the  since the filenames are dynamically
generated.
>
> Is this a bug or am I doing something wrong? I'm using tomcat4.0 for NT
4.0.
>
> THE STACKTRACE:
> 
>
> ApplicationDispatcher[/interestrates]: Servlet.service() for servlet
default
> threw exception
> java.lang.IllegalStateException
> at
>
org.apache.jasper.runtime.ServletResponseWrapperInclude.getOutputStream(Serv
> letResponseWrapperInclude.java:109)
> at
>
org.apache.catalina.servlets.DefaultServlet.serveResource(DefaultServlet.jav
> a:1143)
> at
> org.apache.catalina.servlets.DefaultServlet.doGet(DefaultServlet.java:519)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:740)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
>
org.apache.catalina.core.ApplicationDispatcher.invoke(ApplicationDispatcher.
> java:679)
> at
>
org.apache.catalina.core.ApplicationDispatcher.doInclude(ApplicationDispatch
> er.java:570)
> at
>
org.apache.catalina.core.ApplicationDispatcher.include(ApplicationDispatcher
> .java:493)
> at
>
org.apache.jasper.runtime.JspRuntimeLibrary.include(JspRuntimeLibrary.java:8
> 18)
> at org.apache.jsp.test$jsp._jspService(test$jsp.java:67)
> at org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:107)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
>
org.apache.jasper.servlet.JspServlet$JspServletWrapper.service(JspServlet.ja
> va:202)
> at
org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:382)
> at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:474)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
>
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(Application
> FilterChain.java:247)
> at
>
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterCh
> ain.java:193)
> at
>
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.ja
> va:243)
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 66)
> at
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> at
>
org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.ja
> va:201)
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 66)
> at
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> at
> org.apache.catalina.core.StandardContext.invoke(StandardContext.java:2344)
> at
>
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:164
> )
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 66)
> at
>
org.apache.catalina.valves.ErrorDispatcherValve.invoke(ErrorDispatcherValve.
> java:170)
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 64)
> at
>
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:170
> )
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 64)
> at
>
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:472)
> at org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
> at
>
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java
> :163)
> at
>
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline.java:5
> 66)
> at
>
org.apache.catalina.core.StandardPipeline.invoke(Standar

Re: Unable to start tomcat 4

[Jeff Kilbride]

I don't think Tomcat 4 will run with jdk1.1.8. Try upgrading to jdk1.2 or
later.

--jeff

- Original Message -
From: "Emil Diego" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, December 07, 2001 11:29 AM
Subject: Unable to start tomcat 4


> Hello all,
>
> I have installed Tomcat 4.0.1 onto a Linux box.  I am having trouble
> getting it started.  When i type tomcat4 start, I get the following
> response.
>
> Using CLASSPATH:/var/tomcat4/bin/.bootstrap.jar
> Using CATALINA_BASE:/var/tomcat4
> Using CATALINA_HOME:/var/tomcat4
> Using JAVA_HOME:/usr/jdk1.1.8/
>
> then when i  look at the catalina.log file i see the following entry
> Unable to initilize threads: cannot find class java/lang/Thread
>
>
> For some reason it is failing.  If anyone has any ideas, please let me
> know.  Thanx
>
>
> Emil Diego
> [EMAIL PROTECTED]
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: ONE More Time! why does Tomcat 4 work this way

[Jeff Kilbride]

I believe it's in the jakarta-tomcat-connectors module.

--jeff

- Original Message -
From: "Brian Adams" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 1:17 PM
Subject: RE: ONE More Time! why does Tomcat 4 work this way


> clue please.  were is mod_webapp code?  no see cvs module
>
>
>
>
> -Original Message-
> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 3:09 PM
> To: Tomcat Users List
> Subject: Re: ONE More Time! why does Tomcat 4 work this way
>
>
> Sorry, but I forgot to mention that your other option would be to pull the
> latest mod_webapp code from CVS and see if that helps. I just checked the
> dev list and it looks like somebody is working on it -- though I don't
know
> who...
>
> --jeff
>
> - Original Message -
> From: "Brian Adams" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 12:35 PM
> Subject: RE: ONE More Time! why does Tomcat 4 work this way
>
>
> > I think that is what I would like to do after looking WAY TOO CLOSE at
> this.
> >
> > Patrick I know your out there *peer* what do you say?  Let's move to
> mod_jk
> > *point*
> > B
> >
> > -Original Message-
> > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 2:57 PM
> > To: Tomcat Users List
> > Subject: Re: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > Personally, I would drop mod_webapp in favor of mod_jk, if you really
need
> > to run a separate web server. TC 4 was the first release of mod_webapp,
> > while mod_jk has been around much longer -- and is better tested. AFAIK,
> > there was only one guy working on mod_webapp and he announced he was
> taking
> > a long vacation a few weeks ago. I don't know if anybody has picked up
> where
> > he left off. TC 4.0.1 has a configuration line for mod_jk already
included
> > in server.xml.
> >
> > Looking through the list lately, a large percentage of the messages I
see
> > all relate to problems with mod_webapp.
> >
> > Your call...
> >
> > --jeff
> >
> > P.S. -- I've found TC 4 to be faster without the use of Apache. I use a
> > small footprint webserver (i.e. Acme's thttpd or Dan Bernstein's
> publicfile)
> > to serve images from another port (81) and let TC handle all the dynamic
> > stuff -- I don't have any non-dynamic html files on my site...
> >
> >
> > - Original Message -
> > From: "Brian Adams" <[EMAIL PROTECTED]>
> > To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> > Sent: Thursday, December 06, 2001 12:15 PM
> > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > > I tried that and it does not work either.
> > > I noticed that the examples that come with tomcat 4 does not show the
> > images
> > > also!
> > > This is definitely a problem with WebAPP. I can connect to Tomcat
> directly
> > > 8080 and it works fine.  I've seen others on the list server asking
> about
> > > the same problem...
> > > Who is in charge of WebApp developement?  Their List Serve?
> > > B
> > >
> > > -Original Message-
> > > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, December 06, 2001 2:36 PM
> > > To: Tomcat Users List
> > > Subject: Re: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > What about  just "/images/butterfly.jpg"? This assumes that your
webapp
> is
> > > rooted at /bbb -- so you don't need to use that as part of your path.
> This
> > > is what I normally use with mod_jk. Don't know if it will work with
> > > mod_webapp, though.
> > >
> > > --jeff
> > >
> > > - Original Message -
> > > From: "Brian Adams" <[EMAIL PROTECTED]>
> > > To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> > > Sent: Thursday, December 06, 2001 11:12 AM
> > > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > > there must be some sort of  tag being enforced by tomcat 4.0
> > that
> > > > make me define the images directory before I just start using it...
> > > > any hints?
> > > >

Re: ONE More Time! why does Tomcat 4 work this way

[Jeff Kilbride]

Sorry, but I forgot to mention that your other option would be to pull the
latest mod_webapp code from CVS and see if that helps. I just checked the
dev list and it looks like somebody is working on it -- though I don't know
who...

--jeff

- Original Message -
From: "Brian Adams" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 12:35 PM
Subject: RE: ONE More Time! why does Tomcat 4 work this way


> I think that is what I would like to do after looking WAY TOO CLOSE at
this.
>
> Patrick I know your out there *peer* what do you say?  Let's move to
mod_jk
> *point*
> B
>
> -Original Message-
> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 2:57 PM
> To: Tomcat Users List
> Subject: Re: ONE More Time! why does Tomcat 4 work this way
>
>
> Personally, I would drop mod_webapp in favor of mod_jk, if you really need
> to run a separate web server. TC 4 was the first release of mod_webapp,
> while mod_jk has been around much longer -- and is better tested. AFAIK,
> there was only one guy working on mod_webapp and he announced he was
taking
> a long vacation a few weeks ago. I don't know if anybody has picked up
where
> he left off. TC 4.0.1 has a configuration line for mod_jk already included
> in server.xml.
>
> Looking through the list lately, a large percentage of the messages I see
> all relate to problems with mod_webapp.
>
> Your call...
>
> --jeff
>
> P.S. -- I've found TC 4 to be faster without the use of Apache. I use a
> small footprint webserver (i.e. Acme's thttpd or Dan Bernstein's
publicfile)
> to serve images from another port (81) and let TC handle all the dynamic
> stuff -- I don't have any non-dynamic html files on my site...
>
>
> - Original Message -
> From: "Brian Adams" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 12:15 PM
> Subject: RE: ONE More Time! why does Tomcat 4 work this way
>
>
> > I tried that and it does not work either.
> > I noticed that the examples that come with tomcat 4 does not show the
> images
> > also!
> > This is definitely a problem with WebAPP. I can connect to Tomcat
directly
> > 8080 and it works fine.  I've seen others on the list server asking
about
> > the same problem...
> > Who is in charge of WebApp developement?  Their List Serve?
> > B
> >
> > -Original Message-
> > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 2:36 PM
> > To: Tomcat Users List
> > Subject: Re: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > What about  just "/images/butterfly.jpg"? This assumes that your webapp
is
> > rooted at /bbb -- so you don't need to use that as part of your path.
This
> > is what I normally use with mod_jk. Don't know if it will work with
> > mod_webapp, though.
> >
> > --jeff
> >
> > - Original Message -
> > From: "Brian Adams" <[EMAIL PROTECTED]>
> > To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> > Sent: Thursday, December 06, 2001 11:12 AM
> > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > > there must be some sort of  tag being enforced by tomcat 4.0
> that
> > > make me define the images directory before I just start using it...
> > > any hints?
> > >
> > > -Original Message-
> > > From: Carsten Lingemann [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, December 06, 2001 1:31 PM
> > > To: 'Tomcat Users List'
> > > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > have you tried the following:
> > >
> > > 
> > >
> > > Carsten
> > >
> > > -Original Message-
> > > From: Brian Adams [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, December 06, 2001 2:04 PM
> > > To: '[EMAIL PROTECTED]'
> > > Subject: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > I hace Apache with WebApp warping to tomcat
> > > my jsp page has this image:
> > > 
> > > The web-app is named bbb and it has a directory called images.  The
jsp
> > page
> > > WILL NOT SHOW THIS IMAGE!!!
> > >
> > > If I change it to:
> > > http://localhost/bbb/images/butterfly.jpg"

Re: ONE More Time! why does Tomcat 4 work this way

[Jeff Kilbride]

Dude, you're watching this list WAY too closely! You're average response
time is 2.5 minutes...

:)

--jeff

- Original Message -
From: "Brian Adams" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 12:35 PM
Subject: RE: ONE More Time! why does Tomcat 4 work this way


> I think that is what I would like to do after looking WAY TOO CLOSE at
this.
>
> Patrick I know your out there *peer* what do you say?  Let's move to
mod_jk
> *point*
> B
>
> -Original Message-
> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 2:57 PM
> To: Tomcat Users List
> Subject: Re: ONE More Time! why does Tomcat 4 work this way
>
>
> Personally, I would drop mod_webapp in favor of mod_jk, if you really need
> to run a separate web server. TC 4 was the first release of mod_webapp,
> while mod_jk has been around much longer -- and is better tested. AFAIK,
> there was only one guy working on mod_webapp and he announced he was
taking
> a long vacation a few weeks ago. I don't know if anybody has picked up
where
> he left off. TC 4.0.1 has a configuration line for mod_jk already included
> in server.xml.
>
> Looking through the list lately, a large percentage of the messages I see
> all relate to problems with mod_webapp.
>
> Your call...
>
> --jeff
>
> P.S. -- I've found TC 4 to be faster without the use of Apache. I use a
> small footprint webserver (i.e. Acme's thttpd or Dan Bernstein's
publicfile)
> to serve images from another port (81) and let TC handle all the dynamic
> stuff -- I don't have any non-dynamic html files on my site...
>
>
> - Original Message -
> From: "Brian Adams" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 12:15 PM
> Subject: RE: ONE More Time! why does Tomcat 4 work this way
>
>
> > I tried that and it does not work either.
> > I noticed that the examples that come with tomcat 4 does not show the
> images
> > also!
> > This is definitely a problem with WebAPP. I can connect to Tomcat
directly
> > 8080 and it works fine.  I've seen others on the list server asking
about
> > the same problem...
> > Who is in charge of WebApp developement?  Their List Serve?
> > B
> >
> > -Original Message-
> > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 2:36 PM
> > To: Tomcat Users List
> > Subject: Re: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > What about  just "/images/butterfly.jpg"? This assumes that your webapp
is
> > rooted at /bbb -- so you don't need to use that as part of your path.
This
> > is what I normally use with mod_jk. Don't know if it will work with
> > mod_webapp, though.
> >
> > --jeff
> >
> > - Original Message -
> > From: "Brian Adams" <[EMAIL PROTECTED]>
> > To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> > Sent: Thursday, December 06, 2001 11:12 AM
> > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > > there must be some sort of  tag being enforced by tomcat 4.0
> that
> > > make me define the images directory before I just start using it...
> > > any hints?
> > >
> > > -Original Message-
> > > From: Carsten Lingemann [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, December 06, 2001 1:31 PM
> > > To: 'Tomcat Users List'
> > > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > have you tried the following:
> > >
> > > 
> > >
> > > Carsten
> > >
> > > -Original Message-
> > > From: Brian Adams [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, December 06, 2001 2:04 PM
> > > To: '[EMAIL PROTECTED]'
> > > Subject: ONE More Time! why does Tomcat 4 work this way
> > >
> > >
> > > I hace Apache with WebApp warping to tomcat
> > > my jsp page has this image:
> > > 
> > > The web-app is named bbb and it has a directory called images.  The
jsp
> > page
> > > WILL NOT SHOW THIS IMAGE!!!
> > >
> > > If I change it to:
> > > http://localhost/bbb/images/butterfly.jpg";>  It
> works
> > > fine and fast!
> > >
> > > WHAT THE F*&(* is up with my config or Tomcat???
> > 

Re: ONE More Time! why does Tomcat 4 work this way

[Jeff Kilbride]

Personally, I would drop mod_webapp in favor of mod_jk, if you really need
to run a separate web server. TC 4 was the first release of mod_webapp,
while mod_jk has been around much longer -- and is better tested. AFAIK,
there was only one guy working on mod_webapp and he announced he was taking
a long vacation a few weeks ago. I don't know if anybody has picked up where
he left off. TC 4.0.1 has a configuration line for mod_jk already included
in server.xml.

Looking through the list lately, a large percentage of the messages I see
all relate to problems with mod_webapp.

Your call...

--jeff

P.S. -- I've found TC 4 to be faster without the use of Apache. I use a
small footprint webserver (i.e. Acme's thttpd or Dan Bernstein's publicfile)
to serve images from another port (81) and let TC handle all the dynamic
stuff -- I don't have any non-dynamic html files on my site...


- Original Message -
From: "Brian Adams" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 12:15 PM
Subject: RE: ONE More Time! why does Tomcat 4 work this way


> I tried that and it does not work either.
> I noticed that the examples that come with tomcat 4 does not show the
images
> also!
> This is definitely a problem with WebAPP. I can connect to Tomcat directly
> 8080 and it works fine.  I've seen others on the list server asking about
> the same problem...
> Who is in charge of WebApp developement?  Their List Serve?
> B
>
> -Original Message-
> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 2:36 PM
> To: Tomcat Users List
> Subject: Re: ONE More Time! why does Tomcat 4 work this way
>
>
> What about  just "/images/butterfly.jpg"? This assumes that your webapp is
> rooted at /bbb -- so you don't need to use that as part of your path. This
> is what I normally use with mod_jk. Don't know if it will work with
> mod_webapp, though.
>
> --jeff
>
> - Original Message -
> From: "Brian Adams" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 11:12 AM
> Subject: RE: ONE More Time! why does Tomcat 4 work this way
>
>
> > there must be some sort of  tag being enforced by tomcat 4.0
that
> > make me define the images directory before I just start using it...
> > any hints?
> >
> > -Original Message-
> > From: Carsten Lingemann [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 1:31 PM
> > To: 'Tomcat Users List'
> > Subject: RE: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > have you tried the following:
> >
> > 
> >
> > Carsten
> >
> > -Original Message-
> > From: Brian Adams [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 2:04 PM
> > To: '[EMAIL PROTECTED]'
> > Subject: ONE More Time! why does Tomcat 4 work this way
> >
> >
> > I hace Apache with WebApp warping to tomcat
> > my jsp page has this image:
> > 
> > The web-app is named bbb and it has a directory called images.  The jsp
> page
> > WILL NOT SHOW THIS IMAGE!!!
> >
> > If I change it to:
> > http://localhost/bbb/images/butterfly.jpg";>  It
works
> > fine and fast!
> >
> > WHAT THE F*&(* is up with my config or Tomcat???
> >
> > my warp definition is:
> > everything that has a bbb after the url warps to Tomcat
> > WebAppConnection conn warp localhost:8008
> > WebAppDeploy examples conn /examples
> > WebAppDeploy bbb conn /bbb
> >
> > Am I missing something in my web.xml that should say how to handle the
> > images folder?
> >
> > B
> >
> >
> >
> >
> > --
> > To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> > For additional commands: <mailto:[EMAIL PROTECTED]>
> > Troubles with the list: <mailto:[EMAIL PROTECTED]>
> >
> > --
> > To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> > For additional commands: <mailto:[EMAIL PROTECTED]>
> > Troubles with the list: <mailto:[EMAIL PROTECTED]>
> >
> > --
> > To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> > For additional commands: <mailto:[EMAIL PROTECTED]>
> > Troubles with the list: <mailto:[EMAIL PROTECTED]>
> >
>
>
> --
> To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> For additional commands: <mailto:[EMAIL PROTECTED]>
> Troubles with the list: <mailto:[EMAIL PROTECTED]>
>
> --
> To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> For additional commands: <mailto:[EMAIL PROTECTED]>
> Troubles with the list: <mailto:[EMAIL PROTECTED]>
>


--
To unsubscribe:   <mailto:[EMAIL PROTECTED]>
For additional commands: <mailto:[EMAIL PROTECTED]>
Troubles with the list: <mailto:[EMAIL PROTECTED]>

Re: Setting the alocated memory to Tomcat 4.0.1

[Jeff Kilbride]

Hi Niclas,

You should really do this using the environment variable "CATALINA_OPTS".
That's what it's designed for.

Set up an environment variable called "CATALINA_OPTS" and set it's value to
"-Xms -Xmx". The easiest way to do this is by right-clicking on
your "My Computer" icon and then selecting the "Environment" tab.

Thanks,
--jeff

- Original Message -
From: "Niclas Rothman" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 11:30 AM
Subject: SV: Setting the alocated memory to Tomcat 4.0.1


> Okey Bo!
> It works now if I startup the Tomcat throug startup.bat but it still
doesn´t
> work when I try to
> start the beast as a NT service. Any further suggestions?
>
> Niclas
>
> -Oprindelig meddelelse-
> Fra: Bo Xu [mailto:[EMAIL PROTECTED]]
> Sendt: 6. december 2001 19:50
> Til: Tomcat Users List
> Emne: Re: Setting the alocated memory to Tomcat 4.0.1
>
>
> Hi Niclas,
>
>
> I am not sure because I didn't ever use -Xms/-Xmx to customize
catalina.bat,
> I guess there are two reasons:
>
> - did you update all "_RUNJAVA/_STARTJAVA... start" in catalina.bat?
>in the the following(from TC4.0), there are 4
>"_RUNJAVA/_STARTJAVA... start".
>
> :doRun
> if "%2" == "-security" goto doRunSecure
> %_RUNJAVA%
>
%CATALINA_OPTS% -Dcatalina.base="%CATALINA_BASE%" -Dcatalina.home="%CATALINA
> _HOME%" org.apache.catalina.startup.Bootstrap %2 %3 %4 %5 %6 %7 %8 %9
start
> goto cleanup
> :doRunSecure
> %_RUNJAVA%
>
%CATALINA_OPTS% -Djava.security.manager -Djava.security.policy=="%CATALINA_B
>
ASE%/conf/catalina.policy" -Dcatalina.base="%CATALINA_BASE%" -Dcatalina.home
> ="%CATALINA_HOME%" org.apache.catalina.startup.Bootstrap %3 %4 %5 %6 %7 %8
> %9 start
> goto cleanup
>
> :doStart
> if "%2" == "-security" goto doStartSecure
> %_STARTJAVA%
>
%CATALINA_OPTS% -Dcatalina.base="%CATALINA_BASE%" -Dcatalina.home="%CATALINA
> _HOME%" org.apache.catalina.startup.Bootstrap %2 %3 %4 %5 %6 %7 %8 %9
start
> goto cleanup
> :doStartSecure
> echo Using Security Manager
> %_STARTJAVA%
>
%CATALINA_OPTS% -Djava.security.manager -Djava.security.policy=="%CATALINA_B
>
ASE%/conf/catalina.policy" -Dcatalina.base="%CATALINA_BASE%" -Dcatalina.home
> ="%CATALINA_HOME%" org.apache.catalina.startup.Bootstrap %3 %4 %5 %6 %7 %8
> %9 start
> goto cleanup
>
>
> - does anybody know  if -Xms/-Xmx will increase allocated memory in every
>JVM/OS combination? or it only work in some of them?  Thanks in
advance!
>
> Bo
> Dec.06, 2001
>
>
>
>
>
> - Original Message -
> From: "Niclas Rothman" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 1:27 PM
> Subject: SV: Setting the alocated memory to Tomcat 4.0.1
>
>
> Hi Bo!
> Thank´s for your reply, but I seem to have no luck with this.
> Have done a little jsp file that prints out the free memory and total
memory
> of the JVM and doesn´t seem to change. Maybe I´ve missunderstood your
reply,
> I´ve just added -Xms & -Xmx to the end of the line:
> %CATALINA_OPTS% -Dcatalina.base="%CATALINA_BASE%"
> -Dcatalina.home="%CATALINA
> _HOME%" org.apache.catalina.startup.Bootstrap %2 %3 %4 %5 %6 %7 %8
> %9 start
>
> So the result is:
>
> %CATALINA_OPTS% -Dcatalina.base="%CATALINA_BASE%"
> -Dcatalina.home="%CATALINA
>   _HOME%" org.apache.catalina.startup.Bootstrap %2 %3 %4 %5 %6 %7 %8
%9
> start -Xms100663296
> -Xmx134217728
>
>
> What do say is this "approach" right or wrong?
> Best regards Niclas Rothman
> -Oprindelig meddelelse-
> Fra: Bo Xu [mailto:[EMAIL PROTECTED]]
> Sendt: 6. december 2001 19:10
> Til: Tomcat Users List
> Emne: Re: Setting the alocated memory to Tomcat 4.0.1
>
>
> - Original Message -
> From: "Niclas Rothman" <[EMAIL PROTECTED]>
> To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
> Sent: Thursday, December 06, 2001 12:52 PM
> Subject: Setting the alocated memory to Tomcat 4.0.1
>
>
> Hi everybody!
> I´m using Tomcat 4.0.1 and the version with the Windows NT service.
> Does anybody know where to set the size for the memory to allocate to the
> Tomcat (-Xms & -Xmx), can´t find any documentation about this.
> Best reqards
>
> Niclas Rothman
>
>
> I am not sure, I think you can add it in catalina.bat in
CATALINA_HOME/bin,
> for example, update the following:
> *   _STARTJAVA
> *   %_STARTJAVA%
>
%CATALINA_OPTS% -Dcatalina.base="%CATALINA_BASE%" -Dcatalina.home="%CATALINA
> _HOME%" org.apache.catalina.startup.Bootstrap %2 %3 %4 %5 %6 %7 %8
> %9 start
>
>
> and add java -Xmsn/-Xmxn into them.
>
> Bo
> Dec.06, 2001
>
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>
>
> --
> To unsubscribe:   
> For additional commands: 

Re: ONE More Time! why does Tomcat 4 work this way

[Jeff Kilbride]

What about  just "/images/butterfly.jpg"? This assumes that your webapp is
rooted at /bbb -- so you don't need to use that as part of your path. This
is what I normally use with mod_jk. Don't know if it will work with
mod_webapp, though.

--jeff

- Original Message -
From: "Brian Adams" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 11:12 AM
Subject: RE: ONE More Time! why does Tomcat 4 work this way


> there must be some sort of  tag being enforced by tomcat 4.0 that
> make me define the images directory before I just start using it...
> any hints?
>
> -Original Message-
> From: Carsten Lingemann [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 1:31 PM
> To: 'Tomcat Users List'
> Subject: RE: ONE More Time! why does Tomcat 4 work this way
>
>
> have you tried the following:
>
> 
>
> Carsten
>
> -Original Message-
> From: Brian Adams [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, December 06, 2001 2:04 PM
> To: '[EMAIL PROTECTED]'
> Subject: ONE More Time! why does Tomcat 4 work this way
>
>
> I hace Apache with WebApp warping to tomcat
> my jsp page has this image:
> 
> The web-app is named bbb and it has a directory called images.  The jsp
page
> WILL NOT SHOW THIS IMAGE!!!
>
> If I change it to:
> http://localhost/bbb/images/butterfly.jpg";>  It works
> fine and fast!
>
> WHAT THE F*&(* is up with my config or Tomcat???
>
> my warp definition is:
> everything that has a bbb after the url warps to Tomcat
> WebAppConnection conn warp localhost:8008
> WebAppDeploy examples conn /examples
> WebAppDeploy bbb conn /bbb
>
> Am I missing something in my web.xml that should say how to handle the
> images folder?
>
> B
>
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: tomcat configuration error

[Jeff Kilbride]

Hi Larry,

I thought for all 3.2.x versions, you could just replace the old jar files
with the new ones. Is this not true for 3.2.4?

Thanks,
--jeff

- Original Message -
From: "Larry Isaacs" <[EMAIL PROTECTED]>
To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
Sent: Thursday, December 06, 2001 5:55 AM
Subject: RE: tomcat configuration error


> Note that you say you are configuring Tomcat 3.2.4
> but are setting TOMCAT_HOME to tomcat-3.2.1.
>
> If you are copying Tomcat 3.2.4 on top of a
> Tomcat 3.2.1 installation, be aware that this is
> an untested form of installation.  I would recommend
> installing Tomcat 3.2.4 in a new directory and
> updating the configuration and web apps to match
> your 3.2.1 installation.
>
> Normally the "Unable to set CLASSPATH dynamically ...
> Setting your CLASSPATH statically" message is just
> a warning.  However, in Tomcat 3.2.4 this represents
> an error.
>
> Tomcat 3.2.4 includes JAXP 1.1 as the XML parser
> (consisting of jaxp.jar and crimson.jar) where
> Tomcat 3.2.3 and earlier included JAXP 1.0.1
> (consisting of jaxp.jar and parser.jar).  Note that
> with 3.2.4, it is crimson.jar that is found in the
> TOMCAT_HOME/lib, not parser.jar.
>
> There is a bug in the tomcat.bat file where it tries
> to build the CLASSPATH setting staticially, i.e. with
> a bunch of "SET" commands.  This portion of the batch
> file is only used on Win9x systems when a directory
> in the TOMCAT_HOME path isn't a DOS 8.3 name.  This
> portion of tomcat.bat still specifies parser.jar instead
> of crimson.jar.  Change all "parser.jar" references to
> "crimson.jar" in tomcat.bat and you should be okay.
>
> You could also do what the message says and
> manually set your TOMCAT_HOME using DOS 8.3 names,
> (i.e. SET TOMCAT_HOME=C:\TOMCAT~1 ), or make sure
> all directories in the path to Tomcat are DOS
> 8.3 names, (i.e. C:\Jakarta\Tc324 ).  This would
> allow the CLASSPATH to be build automatically
> to include all jars found in TOMCAT_HOME\lib.
>
> Hope this helps.
>
> Cheers,
> Larry
>
> > -Original Message-
> > From: gaurang khatri [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, December 06, 2001 1:45 AM
> > To: [EMAIL PROTECTED]
> > Subject: tomcat configuration error
> >
> >
> > hello all,
> >
> > I am gaurang. I am new to this group. I am trying to
> > configure tomcat 3.2.4. My JAVA_HOME is c:\jdk1.3.1_01
> > and TOMCAT_HOME is tomcat-3.2.1. I have set JAVA_HOME
> > and TOMCAT_HOME in tomcat.bat. I have also tried
> > tomcat~1.4 in TOMCAT_HOME but getting the same result.
> > Now when I am trying to run "tomcat start" or
> > "startup" from the bin directory
> > of tomcat, I am getting following error. I dont know
> > why I am getting this error.
> >
> > --
> > Unable to set CLASSPATH dynamically.
> > Note: To set the CLASSPATH dynamically on Win9x
> > systems
> >   only DOS 8.3 names may be used in TOMCAT_HOME!
> > Setting your CLASSPATH statically.
> >
> > Using CLASSPATH:
> >
> > Starting Tomcat in new window
> > --
> >
> > and the tomcat is not started in new window. I am
> > using the japanese version of windows 98. Can anybody
> > tell me what is wrong with this. It is very urgent.
> > So, please help me.
> >
> > Thanks to all
> > gaurang.
> >
> >
> > __
> > __
> > For Stock Quotes, Finance News, Insurance, Tax Planners,
> > Mutual Funds...
> > Visit http://in.finance.yahoo.com/
> >
> > --
> > To unsubscribe:   
> > For additional commands: 
> > Troubles with the list: 
> >
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Authentication problem...redirected to /null

[Jeff Kilbride]

You're not supposed to be able to reach the login page, except by accessing
a secure page. The container is then responsible for displaying the login
page and sending the user to the correct secure page, once they have been
authenticated. So, rather than having a link to your login page from your
home page, you should have a link to your main welcome page inside your
secure area. Tomcat will then send the user to the login page automatically,
if they haven't been authenticated.

What version of Tomcat are you using? Unfortunately, in TC 3.2.x (possibly
others, but I'm not sure) the container *redirects* the user to the login
page which makes it possible for the user to then bookmark that page -- thus
defeating the idea that they have to access a secure page first. The only
way I found to get around this was to put my login page in a separate
"/login" directory and then put an "index.jsp" file in that directory that
redirects to my secure area. That way, anyone who bookmarked the login page
was handled correctly. I'm not sure if this will work in other versions of
Tomcat, though.

Hope this helps!

--jeff

- Original Message -
From: "John Mikhail" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, December 05, 2001 5:36 PM
Subject: Authentication problem...redirected to /null


> Hello,
>
> I'm wondering if anyone can help me with an issue I'm having with my web
> app.  I have a web application that uses the JDBCRealm and I've defined
> all the roles and what not.  Here's the scenario...
>
> If I try to access a secure page, it will take me to the login page.  I
> login with a valid user and then get redirected back to the secure page
> with no problems now that I'm authenticated.  That's not the problem.
> The problem is I can also login from the home page.  If I log in from
> the home page with the same authenticated user, it tomcat is trying to
> redirect me to /null.  Why is that?  I have a welcome file list
> defined in my web.xml.  If anyone can help, it would be greatly
> appreciated..
>
>
> --
> John Mikhail
> "Codito, Ergo Sum"
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Double check idiom broken - Tomcat uses it ?

[Jeff Kilbride]

If you really want the developers to take a look at this, you should
probably post it to the tomcat-dev list. It's iffy whether or not they will
see it here.

Thanks,
--jeff

- Original Message -
From: "java programmer" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, December 05, 2001 10:44 PM
Subject: Double check idiom broken - Tomcat uses it ?


> Hi all:
>
> We all know that the lazy-double-check idiom doesn't
> apply to Java because of the Java Memory Model (JMM).
>
> That is to say, look at code such as:
>
> Example a)
> // Set by any other thread other than #1
> volatile boolean stop = false;
>
> // Thread #1 runs this as long as
> // stop is false. Only T1 will call this
> // method, so not synchronized. hence
> // broken due to staleness of 'stop'.
> // synch for _visbility_ ALSO.
> void foo() {
>  while (!stop ) {  //... }
> }
>
>
> Example b): The lazy double check idiom
> public static Foo haha = null;
> public static getFoo() {
> if (foo == null ) {
>   sychronized (Foo.class) {
>  if (foo == null )
> foo = new Foo();
>   }
> }
> return foo;
> }
>
> Both examples are *guaranteed* to be incorrect.
> Note, this is the case, *even* though I am using
> 'volatile' for the stop variable. For more on the
> JMM, consult Item #48 in Effective Java (Josh Bloch),
> look at Bill Pughs' page at:
> http://www.cs.umd.edu/~pugh/java/memoryModel/DoubleCheckedLocking.html
> or check out Doug Lea's stuff.
> Well, here is the thing:
>
> Quite idly, and randomly, I was looking at:
>
> org.apache.jasper.servlet.JspServlet
>
> and I found:
>
> 
> outDated = compiler.isOutDated();
>if(!jsw.isInstantiated() || outDated ) {
>   synchronized(jsw){
> outDated = compiler.compile();
> if(!jsw.isInstantiated() || outDated) {
> if( null ==ctxt.getServletClassName() ) {
> 
>
> This is a complex use of double check type
> code and is really hard to analyse because references
> themselves and what they point to can have
> different levels of staleness (according to the JMM).
> So it's a turbo double idiom type usage, possibly
> incorrect.
>
> I just wanted to bring this to the attention of the
> development team and make sure that *someone* has
> really analysed this according to the JMM. (and
> any other code, similar to this).
>
> Best regards,
>
> [EMAIL PROTECTED]
>
>
>
>
> __
> Do You Yahoo!?
> Send your FREE holiday greetings online!
> http://greetings.yahoo.com
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Tomcat 3.2.3 Memory Exception running on HP-UX 11

[Jeff Kilbride]

For Tomcat 3.2.x, the environment variable is TOMCAT_OPTS.

Thanks,
--jeff

- Original Message -
From: "Chris Newland" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, November 29, 2001 8:40 AM
Subject: RE: Tomcat 3.2.3 Memory Exception running on HP-UX 11


> Hi Mike,
>
> If you are not explicitly setting the JVM initial heap size and maximum
heap
> size then they will take the default values which are quite small (64MByte
> max I think?).
>
> If you want the JVM to use more of the system's memory then start it up
> using the switches:
>
> -Xms  -Xmx 
> (This is for the Sun 1.3 JVM, but I expect the HP JVM has an equivalent)
>
> so for 100MB initial heap, 150MB maximum heap you would change your
startup
> script to override the defaults:
>
> java -Xms100M -Xmx150M 
>
> In catalina.sh (Tomcat 4) there is an environment variable CATALINA_OPTS
> that will add extra parameters to the java command so I just set
> CATALINA_OPTS=" -Xms500M -Xmx1000M". There may be a similar option in
Tomcat
> 3.2.3.
>
> Hope this helps,
>
> Chris
>
> -Original Message-
> From: Michael S. Ricker [mailto:[EMAIL PROTECTED]]
> Sent: 29 November 2001 16:01
> To: [EMAIL PROTECTED]
> Subject: Tomcat 3.2.3 Memory Exception running on HP-UX 11
>
>
> Hello,
>
> Attempted to make 5 simultaneous connections to Tomcat 3.2.3 on HP-UX 11
> and received the following exception. Tomcat is being run with HP Java
> 1.3 without any specific command line parameters. The HP server had
> 200mb real/300mb virtual memory free when monitoring with the top command.
>
> Any suggestions to prevent this exception?
>
> 2001-11-29 16:39:54 - ThreadPool: Caught exception executing
> org.apache.tomcat.service.TcpWorkerThread@dc81fb1e, terminating thread -
> java.lang.OutOfMemoryError: can't create another thread
> at java.lang.Thread.start(Native Method)
> at
> org.apache.tomcat.util.ThreadPool$ControlRunnable.(ThreadPool.java,
> Compiled Code)at
> org.apache.tomcat.util.ThreadPool.openThreads(ThreadPool.java, Compiled
> Code)
> at org.apache.tomcat.util.ThreadPool.runIt(ThreadPool.java, Compiled
> Code)
> at
> org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoint.java,
> Compiled Code)
> at
> org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPool.java,
> Compiled Code)
> at java.lang.Thread.run(Unknown Source)
>
> Mike
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Tomcat 4.01 classloader - HELP!

[Jeff Kilbride]

Hi Chris,

I don't know the answer to this, but you may be better off posting it to the
tomcat-dev list.

Thanks,
--jeff

- Original Message -
From: "Chris Malley" <[EMAIL PROTECTED]>
To: "tomcat-user" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Thursday, November 29, 2001 8:31 AM
Subject: Tomcat 4.01 classloader - HELP!


> I've posted this query a couple of times with no response.
> (See subject "Tomcat 4.01 classloader problem?".)
> One last try before I give up on Tomcat 4.01...
>
> I'm experiencing a problem with SOAP message-style services
> when using Tomcat 4.01. Folks on the soap-user mailing list
> suggested that this may be a classloader problem in Tomcat 4.01.
>
> When trying to access any message-style SOAP service, my client
> receives a "no signature match" fault.  I have no problem with
> RPC-style SOAP services, the same message-style SOAP services
> work fine with Tomcat 3.2.3, and the signature of the my
> message-style services is of the correct form; ie:
>
>   public void serviceName( Envelope env, SOAPContext req, SOAPContext
> res )
> throws IOException, MessagingException;
>
> What makes me think that this is a classloader problem is
> that I can make the fault go away by removing
> $CATALINA_HOME/webapps/soap/WEB-INF/classes/org/.
> But I don't understand why this makes the fault go away,
> and it's clearly an unacceptable hack.
>
> Is anyone else experiencing this problem?
>
> Complete details, source example, and steps to reproduce
> were in my previous post, which I can post again if someone
> is willing to look at this.
>
> Thanks,
>
> -Chris
>
> --
> Chris Malley
> PixelZoom, Inc. Voice: +1.303.494.8849
> 835 Orman Drive EMail: [EMAIL PROTECTED]
> Boulder CO 80303-2616
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: benchmarking Tomcat4?

[Jeff Kilbride]

Not free, but you can get a free trial:

http://www.webperformanceinc.com/

Thanks,
--jeff

- Original Message - 
From: "Matt Egyhazy" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, November 29, 2001 5:52 AM
Subject: Re: benchmarking Tomcat4?


> http://jakarta.apache.org/jmeter/index.html
> 
> matt
> - Original Message - 
> From: "J. Eric Smith" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, November 27, 2001 5:06 PM
> Subject: benchmarking Tomcat4?
> 
> 
> > Has anyone tried benchmarking Tomcat4 versus some of the commercial
> > engines like Websphere or Weblogic?  If so, what did you find out?
> > Also, I'd like to bench my setups to get a feel for where they top out.
> > Any suggestions on what to use?  We've used the  Microsoft web
> > stress tool up to this point in informal testing of Weblogic and such
> > and it's been quite good on Weblogic, but the results it gives using
> > Tomcat4 are not repeatable and are very ambiguous.  Anything out there
> > better that's free?
> >  
> > J. Eric Smith
> > [EMAIL PROTECTED]
> >  
> >  
> > 
> 
> 
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
> 


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: CVS

[Jeff Kilbride]

Anybody interested in CVS should take a look at the online version of Karl
Fogel's book:

cvsbook.red-bean.com

It helped me get up and running with CVS in about a day.

Thanks,
--jeff

- Original Message -
From: "Ralph Einfeldt" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Tuesday, November 20, 2001 7:54 AM
Subject: AW: CVS


> See below:
>
> > -Ursprüngliche Nachricht-
> > Von: Laurent Michenaud [mailto:[EMAIL PROTECTED]]
> > Gesendet: Dienstag, 20. November 2001 16:12
> > An: Tomcat Users List
> > Betreff: RE: CVS
> >
> >
> > We prefer to use cvs rather than Ms SafeSource.
> >
> > I was asking still questions :
> > - What's better ? a repository for each project or a module for
> > each project ?
>
> We don't use either, the projects are simple sub directories under
> one root in the repository.
>
> > - Is it a good idea to use CVS for binary files ? i was thinking
> > about class files, and all the jpeg/gif files...
>
> In general yes. But I would just put class files in the repository
> if they are not build from sources. And make shure that you
> use/configure
> cvs correct to handle binaries. (Otherwise you might expierence some
> surprises.)
>
> > - What would do the site construction script ? pre-compile the jsp and
> > the servlet maybe ? Have u got examples of script that i could see ?
> Checkout all needed files for a given tag.
> Compile the java classes, make a jar of them, fill a test database.
>
> In our environment we have several files that are the same over the
> project that are placed outside the project directory. Our installation
> script copies these files in the deployment site where our webserver
> runs
> and mixes them withe project specific files.
>
> We have automated the setup of a webserver, so that we can setup a new
> instance of the same project in few minutes once we have defined all
> configration parameters. (Even the initial setup of new project
> doesn't take much longer)
>
> This we use to build the web server up to several times a day.
>
> The examples won't help you much because they still work with good old
> JServ.
>
> > - I think we will need differents branches :
> > - one for each stable release
> > - one for a beta developpement but fonctionnal, few bugs.
> > - one for an alpha developpment, used to backup the progress
>
> I wouldn't do to much branches. Just use Tags to mark this versions.
>
> > works( i am hesitating
> > here with a branch for each developper ).
>
> A clear NO! to this approach.
>
> > For example, a developper that has not finished a work at the
> end
> > of the day will update the alpha branch.
>
> If a developer isn't ready, the he shouldn't checkin. So the your
> alpha branch is just the developers workplace in my world. (What do you
> win
> with this checkin ?)
>
> > A developper will update the beta branch when he thinks his
> > source
> > is quite ok.
> > The stable release will be built after hard testing of the
> > global
> > application.
>
> > - Last question : which utils do u use for cvs ? there is wincvs, i've
> > seen webcvs too.
> > Are there any others ?
> We use WinCVS, XEmacs and the command line interface
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Does jakarta-tomcat-service work for *nixes?

[Jeff Kilbride]

I know why Java can't do this inherently. I was hoping for some sort of
tricky wrapper in the j-t-s code.

Lots of security reasons to *not* change to a non-root user? You're saying
it's more secure to run Tomcat as root? I would think it would be the other
way around. Can you elaborate?

Thanks,
--jeff

- Original Message -
From: "Randy Layman" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, November 19, 2001 3:54 AM
Subject: RE: Does jakarta-tomcat-service work for *nixes?


>
>
> > -Original Message-
> > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > Sent: Sunday, November 18, 2001 2:41 AM
> > To: [EMAIL PROTECTED]
> > Subject: Does jakarta-tomcat-service work for *nixes?
> >
> >
> > I went through the archives trying to figure out how to run
> > Tomcat (4.0.1)
> > as a non-root user on port 80. I found references to the
> > jakarta-tomcat-service module in the CVS, so I logged in and
> > downloaded it.
> > Unfortunately, there doesn't seem to be much info on how to
> > set it up -- at
> > least, not for a Linux/Unix environment.
> >
> > Does jakarta-tomcat-service work for Linux? It looks like it
> > may only work
> > for WinNT. If it doesn't work for Linux, is there any other method for
> > starting Tomcat non-root on port 80?
>
> No it doesn't work for Linux.  And no, there is no way for Tomcat to run
as
> a non-root user on port 80 - because Tomcat is written in Java and Java
> doesn't provide the setUID/setGID system calls there is no way to capture
> port 80 as root and then change the user to something else (which is how
the
> other major servers do it - C/C++ provides access to the system call).  By
> the way there are lots of security reasons to not do this.
>
> Randy
>
> >
> > Thanks,
> > --jeff
> >
> >
> >
> > --
> > To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> > For additional commands: <mailto:[EMAIL PROTECTED]>
> > Troubles with the list: <mailto:[EMAIL PROTECTED]>
> >
>
> --
> To unsubscribe:   <mailto:[EMAIL PROTECTED]>
> For additional commands: <mailto:[EMAIL PROTECTED]>
> Troubles with the list: <mailto:[EMAIL PROTECTED]>
>


--
To unsubscribe:   <mailto:[EMAIL PROTECTED]>
For additional commands: <mailto:[EMAIL PROTECTED]>
Troubles with the list: <mailto:[EMAIL PROTECTED]>

Re: AW: Java program interferring with Tomcat

[Jeff Kilbride]

Sure it does. Changing the transaction isolation level affects how data is
read from the database. TRANSACTION_READ_COMMITTED waits for a read lock
before reading data and ensures that the data being read is in a consistent
state. Lower isolation levels may allow data to be read that's
inconsistent -- data from a transaction that hasn't been committed, yet, and
therefore may be rolled back. However, that doesn't mean lower isolation
levels are "bad". They can relieve a lot of locking contention on the
database when used with queries that don't need perfectly consistent data.
It just depends on how critical the info you're pulling with this particular
query is.

For example:  are you tallying the number of visitors to your website or are
you doing financial transactions? With the former, reading "dirty" data may
not matter that much. With the latter, it may be critical that you have
consistent data. If your data's not that critical, you may get a lot of
benefit from lowering your transaction isolation level.

Thanks,
--jeff

- Original Message -
From: "David Frankson" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, November 19, 2001 11:42 AM
Subject: Re: AW: Java program interferring with Tomcat


> > Not with SQL Server.  If thread A has a write lock on a table, thread b
> will
> > wait until the write is completed before reading.  If thread A is in a
> > transaction that is long running then a perfectly funcitoning
application
> > can seem to hang.  Its also possible that some form of infinite loop or
> > deadlock in thread A could cause it to never release the transaction,
> making
> > you either kill the application or use Enterprise Manager to kill the
> lock.
>
> Does changing the transaction isolation level affect things?  The tomcat
> application is a high traffic transactional system, and the commandline
app
> is a read-only data export tool that fires off twice a day.  Both are set
to
> TRANSACTION_READ_COMMITTED.
>
> Dave
>
>
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: LinkageError running 4.0.1 on IBM 1.3.0 VM

[Jeff Kilbride]

Hi Robin,

I'm using the IBM JDK on Linux (RH7.1 w/2.4.14 kernel) with TC 4.0.1 without
any problems. The only difference I can see from your setup is that my
JAVA_HOME variable points to the toplevel of the IBM directory structure:

/usr/local/java/IBMJava2-13

not

/usr/local/java/IBMJava2-13/jre

I get this output when I run startup.sh:

---
Using CLASSPATH:
/usr/local/java/jakarta-tomcat-4.0.1/bin/bootstrap.jar:/usr/local/java/IBMJa
va2-13/lib/tools.jar
Using CATALINA_BASE: /usr/local/java/jakarta-tomcat-4.0.1
Using CATALINA_HOME: /usr/local/java/jakarta-tomcat-4.0.1
Using JAVA_HOME: /usr/local/java/IBMJava2-13
---

Notice that the CLASSPATH contains the "tools.jar" file. On yours it
doesn't. Don't know if this will make a difference or not, but I don't see
anything else that's obvious.

Thanks,
--jeff

- Original Message -
From: "robin" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, November 19, 2001 10:09 AM
Subject: LinkageError running 4.0.1 on IBM 1.3.0 VM


Hi,

I'm trying to run Tomcat 4.0.1 under the IBM JVM 1.3.0.  If I start by
unpacking the archive, and then use the startup.sh script, after setting my
path and unsetting my classpath I get the following output from the scripts:

Starting...
Using CLASSPATH: /usr/local/jakarta-tomcat//bin/bootstrap.jar
Using CATALINA_BASE: /usr/local/jakarta-tomcat/
Using CATALINA_HOME: /usr/local/jakarta-tomcat/
Using JAVA_HOME: /usr/local/IBMJava2-13/jre/


This seems healthy enough, but logs/catalina.out contains:

Catalina.start: LifecycleException:  start: :  java.lang.LinkageError: Class
java/net/URL violates loader constraints
LifecycleException:  start: :  java.lang.LinkageError: Class java/net/URL
violates loader constraints
at java.lang.Throwable.(Throwable.java:84)
at java.lang.Exception.(Exception.java:35)
at
org.apache.catalina.LifecycleException.(LifecycleException.java:126)
at
org.apache.catalina.loader.WebappLoader.start(WebappLoader.java:641)

[ etc. etc. ]

Which is decidedly less healthy, and the server won't respond to any
requests.

I tried the same thing using sun's jdk 1.2.2, and got:

Starting...
Using CLASSPATH:
/usr/local/jakarta-tomcat//bin/bootstrap.jar:/usr/local/jdk1.2.2//lib/tools.
jar
Using CATALINA_BASE: /usr/local/jakarta-tomcat/
Using CATALINA_HOME: /usr/local/jakarta-tomcat/
Using JAVA_HOME: /usr/local/jdk1.2.2/


from the script, and a working server.

The platform is linux 2.2.12-20 on a PIII.

After I hit this problem, I tried building from source under the IBM jvm,
mainly to see whether it would throw up anything useful.  I hit on a
'sealing violation' right at the beginning of the build, in ANT, and this
reproduced on the sun jvm .  I am aware that this means that there are more
than one jar containing members of a package and at least one of them is
sealed.  I tried clearing the classpath altogether, but got the same
problem.  In any case - all I want is a binary version which works with the
IBM vm.

Am I doing something obviously wrong?

Thanks.

-Robin Barooah



--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Does jakarta-tomcat-service work for *nixes?

[Jeff Kilbride]

I went through the archives trying to figure out how to run Tomcat (4.0.1)
as a non-root user on port 80. I found references to the
jakarta-tomcat-service module in the CVS, so I logged in and downloaded it.
Unfortunately, there doesn't seem to be much info on how to set it up -- at
least, not for a Linux/Unix environment.

Does jakarta-tomcat-service work for Linux? It looks like it may only work
for WinNT. If it doesn't work for Linux, is there any other method for
starting Tomcat non-root on port 80?

Thanks,
--jeff



--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Common Problem with Tomcat/Apache (processer sticks at 97% on linux)

[Jeff Kilbride]

- Original Message -
From: "Brandon Cruz" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Friday, November 16, 2001 7:01 AM
Subject: RE: Common Problem with Tomcat/Apache (processer sticks at 97% on
linux)


> Thank you for the suggestions, so to upgrade to a newer version of tomcat
> 3.2.x, all I have to do is switch jar files in tomcat_home/lib?  Do you
> think that may help with the 97% processor problem?

That's what the Readme file that comes with the distribution says -- and it
worked for me. I remember 3.2.1 (and possibly 3.2.2) having a known runaway
processor problem, but I don't remember what caused it. Something to do with
mapping a servlet incorrectly could cause a runaway, I think. I believe it
was fixed by 3.2.3. As I said, I think 3.2.4 is going to be released very
soon, so you might wait for that distro.

> As far as garbage collection, this condition stays this way, even after 12
> hours.  Would that be an indication of garbage collection or an endless
loop
> somewhere?  If it is in our developers code, then I would expect this to
> reproduce on any machine, development on win2k, or production on
> linux-apache-tomcat.  Still can only find this on the linux-apache
machine.
>
> I will try the upgrade and reading up some information on garbage
> collection.  If anyone else has seen this condition and been able to fix
it,
> please let me know what you did!

I doubt that it's a garbage collection issue. Try upgrading TC first to see
if it resolves your problems.

> Thanks Again!
>
> Brandon

--jeff


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Common Problem with Tomcat/Apache (processer sticks at 97% on linux)

[Jeff Kilbride]

At the very least, upgrade to the latest stable 3.2.x version. 3.2.1 is
quite old and had several security problems. I'm pretty sure they are about
to release 3.2.4, so that might be a good incentive. You don't have to
change any of your config files, just replace the jar files in your
TOMCAT_HOME/lib directory with the new ones.

Thanks,
--jeff

- Original Message -
From: "Brandon Cruz" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, November 15, 2001 5:08 PM
Subject: Common Problem with Tomcat/Apache (processer sticks at 97% on
linux)


> I am using Tomcat 3.2.1 connected to Apache via mod_jk.  For some reason,
> every once in a while, my java (Tomcat) process in linux goes up to 97%
and
> stays there until I try to stop tomcat.  Even though tomcat shuts down,
that
> process still stays all of the way up at 97% and can only be stopped by
> executing the kill command.  Does anyone know a fix for this problem,
> upgrade tomcat, upgrade linux version, upgrade jre, etc.?
>
> I have seen many posts on this subject before, but never a definate
answer.
> I see that many people say check the code, so I am doing that for now.
The
> only strange thing is that I can't figure out exactly what causes this to
> happen, and it can't be duplicated on development machines running tomcat
> stand alone on win2k.
>
> Any help would REALLY be appreciated!!!
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Wrap an HttpServletRequest

[Jeff Kilbride]

Hi Diego,

You can also do this by adding query string data to the URL you are
forwarding to, for example:

-
String myPage = "/path/myPage.jsp";
String myForward = myPage + "?newParam1=value1&newParam2=value2";

RequestDispatcher rd = getServletContext().getRequestDispatcher(myForward);
rd.forward(request, response);
-

This way, your new info is added to the request as additional query string
data. Your existing JSP pages can access the info with the normal
request.getParameter() calls. They don't have to know that the request has
been "wrapped".

Thanks,
--jeff

- Original Message -
From: "Diego del Río" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Wednesday, November 14, 2001 6:39 AM
Subject: Re: Wrap an HttpServletRequest


> Thank you Thomas and Martin,
> our problem is that the target servlet, i.e. the servlet that would
receive
> the 'wrapped request' forwarded by the first servlet (in this case, the
> target
> servlet is a jsp page), shouldn't know that the request is wrapped.
> We have already implemented a lot of jsp pages (servlets at last) that
> aren't
> aware about attributes in the request they process.
> I know that this functionality is provided by 'filters' in the servlet 2.3
> spec.; unfortunately we are using Tomcat 3.2.3, a servlet 2.2 engine.
>
> > Hi Diego,
> >
> > do it like Martin described. Or try out the RequestDispatcher:
> >
> > request.setAttribute(key1, value1);
> > request.setAttribute(key2, value2);
> > [...]
> > RequestDispatcher rd =
> > getServletContext().getRequestDispatcher(myServletUrl);
> > rd.forward(request, response);
> >
> > Thomas
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>


--
To unsubscribe:   
For additional commands: 
Troubles with the list: 

Re: Is there a way to make a default web.xml? --Tomcat 3.x

[Jeff Kilbride]

Hi Rob,

3.2.x still has web.xml in $TOMCAT_HOME/conf. I haven't tried 3.3 yet.

Thanks,
--jeff

- Original Message -
From: "Rob S." <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, August 24, 2001 7:42 AM
Subject: Re: Is there a way to make a default web.xml? --Tomcat 3.x


> Someone correct me if I'm wrong plz =) but I believe the web.xml in
$TOMCAT_HOME/conf is applied to all web apps.  I think I read on the dev
list (from Larry) that this feature was removed in 3.x to increase app
portability ===> that relying on a default web.xml reduces the portability
of an application.
>
> However, this behavior is indeed present in TC4.
>
> - r
>
> On Fri, 24 Aug 2001 09:22:32 -0500 [EMAIL PROTECTED] wrote:
> > Sorry, forgot to mention tomcat version 3.x
> > -Original Message-
> > From: Brandon Cruz [mailto:[EMAIL PROTECTED]]
> > Sent: Friday, August 24, 2001 9:17 AM
> > To: [EMAIL PROTECTED]
> > Subject: Is there a way to make a default web.xml?
> >
> >
> > Is there a way to make a default web.xml file that will work across all
> > contexts/virtual hosts?
> >
> > Brandon
>
>
>

Re: VHosts causing app to load twice

[Jeff Kilbride]

Hi Steve,

I spent a lot of time on this when I first set up 3.2.2. I was going to do a
patch myself, until I found out that the 3.2.x versions were frozen and
wouldn't be adding any new features. (I really consider this a necessity,
though...) However, I decided to bite the bullet and only set up one domain
per webapp.

Looks like the patches I remember were for 3.3 and 4.0 -- nothing for
3.2.x... (coulda sworn!)

http://marc.theaimsgroup.com/?l=tomcat-dev&m=99267441502219&w=2
http://marc.theaimsgroup.com/?l=tomcat-dev&m=99663162526421&w=2

Thanks,
--jeff

- Original Message -
From: "Steve Heard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 20, 2001 1:24 PM
Subject: Re: VHosts causing app to load twice


> Jeff,
>
> You've managed to articulate the problem much better then my original
post!
> I'm afraid as you've said, it is unfortunate that this isn't possible in
> 3.2.
>
> I hunted around for that patch that you thought might exist that allows
> multiple hosts per entry but couldn't find it. Anyone know where it might
> be?
>
> Thanks,
> -Steve
>
>
> on 8/20/01 3:48 PM, Larry Isaacs at [EMAIL PROTECTED] wrote:
>
> > Hi Jeff,
> >
> > Maybe I can increase my understanding here, since I not
> > that experienced with the Apache side.
> >
> > The point of:
> >
> > NameVirtualHost 111.22.33.44
> >
> > 
> > ServerName www.foo.com
> > ...
> > 
> >
> > was that mod_jk would identify the host as "www.foo.com"
> > for both Both "http://111.22.33.44/..."; and
> > "http://www.foo.com/...";.  Thus, Tomcat could match it to a
> > single context identified by "www.foo.com".  I can't claim to
> > understand the NameVirtualHost directive very well, but without
> > it, I believe that mod_jk would identify the host as
> > "111.22.33.44" instead of "www.foo.com" for
> > "httpd://111.22.33.44/...".
> >
> > Obviously, I'm coming at this from the point of view of
> > keeping mod_jk and Tomat happy, and not necessarily what
> > is "normal" for Apache.
> >
> > By the way, ApacheConfig in Tomcat 3.3 also supports the
> > following for "ServerAlias":
> >
> > server.xml:
> > 
> > 
> > 
> > 
> >
> > which would generate conf/auto/mod_jk.conf:
> >
> > NameVirtualHost 111.22.33.44
> >
> > 
> > ServerName www.foo.com
> > ServerAlias www.bar.com
> > ...
> > 
> >
> > Cheers,
> > Larry
> >
> >> -Original Message-
> >> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> >> Sent: Monday, August 20, 2001 3:21 PM
> >> To: [EMAIL PROTECTED]
> >> Subject: Re: VHosts causing app to load twice
> >>
> >>
> >> Hi Larry,
> >>
> >> It doesn't really make sense to set up Name-based virtual
> >> hosting this way.
> >> In essence, your saying all requests for 111.22.33.44 should
> >> go to the same
> >> place as all requests for www.foo.com (in httpd.conf). In
> >> that case, I think
> >> you should stick with IP-based vhosts. Your Apache config
> >> would then be:
> >>
> >> 
> >> 
> >> 
> >> 
> >>
> >> With this setup, the "host" header is ignored and all requests to
> >> 111.22.33.44, no matter what domain they come in on, will go
> >> to the right
> >> place -- from Apache's standpoint, I don't know if it's
> >> handled correctly in
> >> Tomcat. You should be able to set up your "Host" directive in
> >> server.xml to
> >> catch the IP Address and make it work that way, though.
> >>
> >> I think the real problem with Name-based vhosts comes when
> >> you're trying to
> >> point more than one domain name at the same Tomcat webapp:
> >>
> >> NameVirtualHost 111.22.33.44
> >>
> >> 
> >> ServerName "foo.com"
> >> ServerAlias "www.foo.com"
> >> ...
> >> 
> >>
> >> 
> >> ServerName "bar.com"
> >> ServerAlias "www.bar.com"
> >> ...
> >> 
> >>
> >> Now, if you want foo.com, www.foo.com, bar.com, and
> >> www.bar.com to all point
> >> at the same webapp, there's no way to do it without having
> >> four different
> >> insta

Re: VHosts causing app to load twice

[Jeff Kilbride]

Hi Larry,

My comments are interspersed...   :)

- Original Message -
From: "Larry Isaacs" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 20, 2001 12:48 PM
Subject: RE: VHosts causing app to load twice


> Hi Jeff,
>
> Maybe I can increase my understanding here, since I not
> that experienced with the Apache side.
>
> The point of:
>
> NameVirtualHost 111.22.33.44
>
> 
>ServerName www.foo.com
> ...
> 
>
> was that mod_jk would identify the host as "www.foo.com"
> for both Both "http://111.22.33.44/..."; and
> "http://www.foo.com/...";.  Thus, Tomcat could match it to a
> single context identified by "www.foo.com".  I can't claim to
> understand the NameVirtualHost directive very well, but without
> it, I believe that mod_jk would identify the host as
> "111.22.33.44" instead of "www.foo.com" for
> "httpd://111.22.33.44/...".

Ok, now I see the problem that you're describing. However, I still think
this can be solved with an IP-based vhost where the config would be:


  ServerName "www.foo.com"
  


Notice there's no NameVirtualHost directive. Here we're telling Apache that
every request that comes in on this IPAddress will have it's ServerName set
to www.foo.com. The only time you would need the NameVirtualHost directive
is if you want to set up more domains with the same IP. For example:

NameVirtualHost 111.22.33.44


  ServerName "www.foo.com"
  .



  ServerName "www.bar.com"
  .


Now, Apache will use the "host" header supplied by the browser to figure out
which vhost to send the request to. If no "host" header is supplied, I
believe it defaults to the first vhost in the list -- problematic for
browsers that don't supply a "host" header.

I don't think there's that big of a difference between the way you had it
and my first config above, just saves a bit of typing in httpd.conf and
looks cleaner from an Apache point of view. The results are probably the
same...

> Obviously, I'm coming at this from the point of view of
> keeping mod_jk and Tomat happy, and not necessarily what
> is "normal" for Apache.
>
> By the way, ApacheConfig in Tomcat 3.3 also supports the
> following for "ServerAlias":
>
> server.xml:
>   
> 
> 
>   
>
> which would generate conf/auto/mod_jk.conf:
>
>   NameVirtualHost 111.22.33.44
>
>   
> ServerName www.foo.com
> ServerAlias www.bar.com
> ...
>   
>
> Cheers,
> Larry

Cool! This is what I was looking for in 3.2.x. Any hopes of migrating this
back into 3.2.4 or later? Kinda silly not to have this functionality. Also,
can you have multiple "Alias" directives? (I'm not even sure if you can do
that in Apache...)

Thanks,
--jeff

Re: VHosts causing app to load twice

[Jeff Kilbride]

Hi Larry,

It doesn't really make sense to set up Name-based virtual hosting this way.
In essence, your saying all requests for 111.22.33.44 should go to the same
place as all requests for www.foo.com (in httpd.conf). In that case, I think
you should stick with IP-based vhosts. Your Apache config would then be:


  
  


With this setup, the "host" header is ignored and all requests to
111.22.33.44, no matter what domain they come in on, will go to the right
place -- from Apache's standpoint, I don't know if it's handled correctly in
Tomcat. You should be able to set up your "Host" directive in server.xml to
catch the IP Address and make it work that way, though.

I think the real problem with Name-based vhosts comes when you're trying to
point more than one domain name at the same Tomcat webapp:

NameVirtualHost 111.22.33.44


ServerName "foo.com"
ServerAlias "www.foo.com"
...



ServerName "bar.com"
ServerAlias "www.bar.com"
...


Now, if you want foo.com, www.foo.com, bar.com, and www.bar.com to all point
at the same webapp, there's no way to do it without having four different
instances of your webapp -- at least not in TC 3.2.x. I don't know about
3.3, but I think 4.0 has an "Alias" directive. I'm not sure if it's actually
implemented, though.

It may be kind of extreme to have different vhosts pointing at the same
webapp, however I think it's pretty common to want foo.com and www.foo.com
to point to the same place. It's really unfortunate that Tomcat (3.2.x)
doesn't handle this without having to instantiate two different webapps. I'd
be interested in fixing this in 3.2.x, if development wasn't already
frozen...

Thanks,
--jeff

- Original Message -
From: "Larry Isaacs" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 20, 2001 11:23 AM
Subject: RE: VHosts causing app to load twice


> In Tomcat 3.3 I have made changes to ApacheConfig.java to try
> to address this.  I think with Tomcat 3.2.2 you will need to
> do the Apache part manually, which it seems you may already
> be doing.
>
> I haven't done much with Vhosts on Apache beyond my tests
> while playing with ApacheConfig.  If I am in error, someone
> let me know. It is my understanding that what you want is:
>
> server.xml
>  
>
>  
>
> httpd.conf or an include:
>   NameVirtualHost 111.22.33.44
>
>   
> ServerName www.foo.com
> ...
>   
>
> For me, this works under Tomcat 3.3.  Both "http://111.22.33.44/...";
> and "http://www.foo.com/..."; come over to Tomcat as "www.foo.com".
> I haven't tried this with Tomcat 3.2.x.  I also haven't tried this
> with *nix, just Win2k and Win98.
>
> One additional note.  I had lots of trouble dealing with the "root"
> contexts.  Your httpd.conf below says the "root" context (i.e.
> DocumentRoot) is at "/usr/local/apache/htdocs" and Tomcat thinks
> it is at "/usr/local/apache/servlets".  I think this situation is
> ripe for problems.  Not knowing your application, it is hard to say
> what the best approach would be.
>
> Larry
>
>
> > -Original Message-
> > From: Steve Heard [mailto:[EMAIL PROTECTED]]
> > Sent: Monday, August 20, 2001 1:44 PM
> > To: [EMAIL PROTECTED]
> > Subject: Re: VHosts causing app to load twice
> >
> >
> > 3.2.2
> >
> >
> >
> >
> > on 8/20/01 12:53 PM, Larry Isaacs at [EMAIL PROTECTED] wrote:
> >
> > > Which version Tomcat are you using?  I have tried to deal with
> > > this issue in Tomcat 3.3.
> > >
> > > Larry
> > >
> > >> -Original Message-
> > >> From: Steve Heard [mailto:[EMAIL PROTECTED]]
> > >> Sent: Monday, August 20, 2001 12:39 PM
> > >> To: [EMAIL PROTECTED]
> > >> Subject: VHosts causing app to load twice
> > >>
> > >>
> > >> I have a servlet based application that needs to be accessed
> > >> using both the
> > >> the server IP address and the server domain name.
> > >>
> > >> The problem is that because I have two  entries for the same
> > >> application tomcat "loads" them twice. Not only does this
> > >> leave me with two
> > >> instances of the servlets, but the ones I have  marked to
> > >> load on init have
> > >> their code called twice which interferes with itself.
> > >>
> > >> I have tried every combination of VirtualHosts I can think of
> > >> and always run
> > >> into the issue of in order for Tomcat to recognize that it
> > >> should handle the
> > >> request to the servlet I have to add a  entry in
> > >> server.xml. If I
> > >> leave out the entry for the IP address Apache passes the
> > >> request along to
> > >> tomcat fine, but then Tomcat doesn't recognize it. Likewise
> > >> for the domain
> > >> name.
> > >>
> > >> Any thoughts or help? I have included snippets of what I
> > think are the
> > >> relevant config files and how they should look.
> > >>
> > >> -Steve
> > >>
> > >> --
> > >> From http.conf:
> > >>
> > >> ...
> > >>
> > >> NameVirtualHost 111.22.33.44
> > >>
> > >> 
> > >> ServerName 111.22.33.44
> > >> Err

Re: VHosts causing app to load twice

[Jeff Kilbride]

Hi Steve,

I ran into the same problem when trying to configure "mydomain.com" and
"www.mydomain.com" under the same VHost. I was using the "ServerAlias"
directive in Apache. Tomcat (3.2.x) needed two Host directives to deal with
both names, which would give me two instances of my webapp. I finally gave
up and only implemented the "www.mydomain.com"

I'm under the impression that TC 4.0b7 has the ability to add aliases much
like Apache using an "Alias" directive in server.xml. You might want to
check that out. Also, someone posted a patch for 3.2.x to the tomcat-dev
list a month or so ago that allowed you to use a list of domain names in
your "Host" directive, like:



You might want to search the dev archives and try this patch. I haven't
tried it yet, so I don't know if it works.

Thanks,
--jeff

- Original Message -
From: "Steve Heard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 20, 2001 10:43 AM
Subject: Re: VHosts causing app to load twice


> Jeff,
> I am indeed running other hosts on this guy, just didn't include them in
the
> snippet! Sorry about that.
> Thanks,
> steve
>
>
>
> on 8/20/01 1:07 PM, Jeff Kilbride at [EMAIL PROTECTED] wrote:
>
> > Hi Steve,
> >
> > If you must be able to pull this site up by it's IP Address, then you
must
> > not be VHosting any other domains on this IP. Is that correct?
> >
> > If so, you should be using IP-based VHosting instead of Name-based.
IP-based
> > VHosting will associate one IP with one domain name. Tomcat shouldn't
have a
> > problem once it's set up this way.
> >
> > Check out:
> >
> > http://httpd.apache.org/docs/vhosts/ip-based.html
> >
> > Thanks,
> > --jeff
> >
> > - Original Message -
> > From: "Steve Heard" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Monday, August 20, 2001 9:39 AM
> > Subject: VHosts causing app to load twice
> >
> >
> >> I have a servlet based application that needs to be accessed using both
> > the
> >> the server IP address and the server domain name.
> >>
> >> The problem is that because I have two  entries for the same
> >> application tomcat "loads" them twice. Not only does this leave me with
> > two
> >> instances of the servlets, but the ones I have  marked to load on init
> > have
> >> their code called twice which interferes with itself.
> >>
> >> I have tried every combination of VirtualHosts I can think of and
always
> > run
> >> into the issue of in order for Tomcat to recognize that it should
handle
> > the
> >> request to the servlet I have to add a  entry in server.xml. If I
> >> leave out the entry for the IP address Apache passes the request along
to
> >> tomcat fine, but then Tomcat doesn't recognize it. Likewise for the
domain
> >> name.
> >>
> >> Any thoughts or help? I have included snippets of what I think are the
> >> relevant config files and how they should look.
> >>
> >> -Steve
> >>
> >> --
> >>> From http.conf:
> >>
> >> ...
> >>
> >> NameVirtualHost 111.22.33.44
> >>
> >> 
> >> ServerName 111.22.33.44
> >> ErrorLog /usr/local/apache/logs/error_log
> >> DocumentRoot /usr/local/apache/htdocs
> >>
> >> JkMount /*.servlet ajp13
> >> 
> >>
> >> 
> >> ServerName www.foo.com
> >> ErrorLog /usr/local/apache/logs/error_log
> >> DocumentRoot /usr/local/apache/htdocs
> >>
> >> JkMount /*.servlet ajp13
> >> 
> >> --
> >>
> >> --
> >>> From server.xml:
> >>
> >> ...
> >>
> >> 
> >> 
> >> 
> >>
> >>
> >> 
> >> 
> >> 
> >>
> >> ...
> >> --
> >>
> >
> >
>

Re: VHosts causing app to load twice

[Jeff Kilbride]

Hi Steve,

If you must be able to pull this site up by it's IP Address, then you must
not be VHosting any other domains on this IP. Is that correct?

If so, you should be using IP-based VHosting instead of Name-based. IP-based
VHosting will associate one IP with one domain name. Tomcat shouldn't have a
problem once it's set up this way.

Check out:

http://httpd.apache.org/docs/vhosts/ip-based.html

Thanks,
--jeff

- Original Message -
From: "Steve Heard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 20, 2001 9:39 AM
Subject: VHosts causing app to load twice


> I have a servlet based application that needs to be accessed using both
the
> the server IP address and the server domain name.
>
> The problem is that because I have two  entries for the same
> application tomcat "loads" them twice. Not only does this leave me with
two
> instances of the servlets, but the ones I have  marked to load on init
have
> their code called twice which interferes with itself.
>
> I have tried every combination of VirtualHosts I can think of and always
run
> into the issue of in order for Tomcat to recognize that it should handle
the
> request to the servlet I have to add a  entry in server.xml. If I
> leave out the entry for the IP address Apache passes the request along to
> tomcat fine, but then Tomcat doesn't recognize it. Likewise for the domain
> name.
>
> Any thoughts or help? I have included snippets of what I think are the
> relevant config files and how they should look.
>
> -Steve
>
> --
> >From http.conf:
>
> ...
>
> NameVirtualHost 111.22.33.44
>
> 
>   ServerName 111.22.33.44
>   ErrorLog /usr/local/apache/logs/error_log
>   DocumentRoot /usr/local/apache/htdocs
>
>   JkMount /*.servlet ajp13
> 
>
> 
>   ServerName www.foo.com
>   ErrorLog /usr/local/apache/logs/error_log
>   DocumentRoot /usr/local/apache/htdocs
>
>   JkMount /*.servlet ajp13
> 
> --
>
> --
> >From server.xml:
>
> ...
>
> 
>   
> 
>
>
> 
>   
>  
>
> ...
> --
>

Re: db connection denied access

[Jeff Kilbride]

You need to read chapters 4.2 and 4.3 of the MySQL manual:

http://www.mysql.com/doc/P/r/Privilege_system.html
http://www.mysql.com/doc/U/s/User_Account_Management.html

MySQL has a somewhat strange privilege system for granting access to the
database. But it's very flexible, once you learn it.

If you're going to be using MySQL, do yourself a favor and buy this book:

http://www.amazon.com/exec/obidos/ASIN/0735709211/qid=997901246/sr=2-2/107-4
535935-5049369

It's invaluable.

--jeff

- Original Message -
From: "Richard Draucker" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 15, 2001 8:04 AM
Subject: db connection denied access


> This is odd and I'm stumped.  I have a simple db connection class in a jar
> under ~app/WEB-INF/lib.  I have a servlet under ~app/WEB-INF/classes that
> imports this connection class.  The mm.mysql driver jar also resides in
this
> WEB-INF/lib.
> The servlet accesses the connection class just fine.  But, when the
> connection class attempts a connection to the database I get the error:
>
> java.sql.SQLException: Server configuration denies access to data source
>
> However, if I cut 'n paste the code from the connection class into the
> servlet I have no problem with the connection or data retrieval.
>
> FYI:
> RH7.1 new installation
> jdk 1.3.1
> Tomcat 3.2.1
> MySQL 3.2.2
>
>
> --
> Richard Draucker [EMAIL PROTECTED]
> Protected-Data.Com www.protected-data.com
> Remote Data Support For Web Developers
>

Re: JDBC

[Jeff Kilbride]

No. You need to download the mysql drivers from

http://mmmysql.sourceforge.net/

Put the jar file in your classpath and you're set.

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, August 09, 2001 1:47 PM
Subject: JDBC


> Does Tomcat have all the necessary JDBC drivers installed to successfully
> connect to a mysql database?  I will be installing Tomcat on Linux
Mandrake
> 8with Apache shortly.  And I will be developing JSP programs that use
MySql.
>
>
> Thanks,
> Michael
> Johnson
>
> -
> This message was sent using Endymion MailMan.
> http://www.endymion.com/products/mailman/
>
>

Re: Cookie Exception: java.lang.IllegalArgumentException

[Jeff Kilbride]

Hi Larry,

I can put some logging statements in and see if I can find out more about
what's going on. However, I think the call creating the new cookie should at
least be wrapped in a try/catch to prevent the exception from blowing that
particular request. I'll do that and submit it to the Dev list.

Thanks,
--jeff

- Original Message -
From: "Larry Isaacs" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, August 09, 2001 7:05 AM
Subject: RE: Cookie Exception: java.lang.IllegalArgumentException


> This bug appears more than once in Bugzilla.  So far, attempts
> to duplicate it haven't been successful, so it has never
> been tracked down.
>
> Cookie handling has been rewritten in Tomcat 3.3 and the
> bugs have been "resolved" as being fixed in 3.3.
>
> If you are interested, I can try to supply advice about
> preparing a slightly customized Tomcat 3.2.3 that
> dumps additional information.  Maybe we can get a
> clue about what is going wrong.
>
> I'd offer to do more, but my hands are full with Tomcat 3.3.
>
> Cheers,
> Larry
>
> > -Original Message-
> > From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, August 09, 2001 4:27 AM
> > To: [EMAIL PROTECTED]
> > Subject: Cookie Exception: java.lang.IllegalArgumentException
> >
> >
> > Has anyone seen this type of Exception before? I'm running:
> >
> > RH 6.2 (2.4.7 kernel)
> > IBMJava2_13
> > Tomcat 3.2.3
> > Apache 1.3.19
> > mod_jk
> >
> > This is a light to moderately loaded webserver (50,000
> > hits/day) and I'm
> > getting 20 or 30 of these in my logs each day. I am also
> > seeing errors where
> > the message says "Cookie name path is a reserved token". From tracing
> > through the code, it seems that
> > org.apache.tomcat.util.RequestUtil.processCookies is trying
> > to create a
> > cookie named "Expires" or "path". (It takes the cookie header string,
> > tokenizes it, and creates Cookie objects based on name=value
> > pairs.) I've
> > never seen "Expires" or "path" in a cookie header. Is this
> > happening because
> > the browser is sending a bad cookie header, or is it some
> > other problem?
> >
> > Thanks,
> > --jeff
> >
> > 
> > 2001-08-08 20:40:15 - Ctx( www..com: ): Exception in: R(  +
> > /servlet/x + null) - java.lang.IllegalArgumentException:
> > Cookie name
> > Expires is a reserved token
> > at javax.servlet.http.Cookie.(Cookie.java(Compiled Code))
> > at
> > org.apache.tomcat.util.RequestUtil.processCookies(RequestUtil.
> > java(Compiled
> > Code))
> > at
> > org.apache.tomcat.core.RequestImpl.getCookieCount(RequestImpl.
> > java(Compiled
> > Code))
> > at
> > org.apache.tomcat.session.StandardSessionInterceptor.requestMa
> > p(StandardSess
> > ionInterceptor.java(Compiled Code))
> > at
> > org.apache.tomcat.core.ContextManager.processRequest(ContextMa
> > nager.java(Com
> > piled Code))
> > at
> > org.apache.tomcat.core.ContextManager.internalService(ContextM
> > anager.java(Co
> > mpiled Code))
> > at
> > org.apache.tomcat.core.ContextManager.service(ContextManager.j
> > ava(Compiled
> > Code))
> > at
> > org.apache.tomcat.service.connector.Ajp13ConnectionHandler.pro
> > cessConnection
> > (Ajp13ConnectionHandler.java:160)
> > at
> > org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoin
> > t.java:416)
> > at
> > org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPo
> > ol.java:501)
> > at java.lang.Thread.run(Thread.java:498)
> >
> >
>

Cookie Exception: java.lang.IllegalArgumentException

[Jeff Kilbride]

Has anyone seen this type of Exception before? I'm running:

RH 6.2 (2.4.7 kernel)
IBMJava2_13
Tomcat 3.2.3
Apache 1.3.19
mod_jk

This is a light to moderately loaded webserver (50,000 hits/day) and I'm
getting 20 or 30 of these in my logs each day. I am also seeing errors where
the message says "Cookie name path is a reserved token". From tracing
through the code, it seems that
org.apache.tomcat.util.RequestUtil.processCookies is trying to create a
cookie named "Expires" or "path". (It takes the cookie header string,
tokenizes it, and creates Cookie objects based on name=value pairs.) I've
never seen "Expires" or "path" in a cookie header. Is this happening because
the browser is sending a bad cookie header, or is it some other problem?

Thanks,
--jeff


2001-08-08 20:40:15 - Ctx( www..com: ): Exception in: R(  +
/servlet/x + null) - java.lang.IllegalArgumentException: Cookie name
Expires is a reserved token
at javax.servlet.http.Cookie.(Cookie.java(Compiled Code))
at
org.apache.tomcat.util.RequestUtil.processCookies(RequestUtil.java(Compiled
Code))
at
org.apache.tomcat.core.RequestImpl.getCookieCount(RequestImpl.java(Compiled
Code))
at
org.apache.tomcat.session.StandardSessionInterceptor.requestMap(StandardSess
ionInterceptor.java(Compiled Code))
at
org.apache.tomcat.core.ContextManager.processRequest(ContextManager.java(Com
piled Code))
at
org.apache.tomcat.core.ContextManager.internalService(ContextManager.java(Co
mpiled Code))
at
org.apache.tomcat.core.ContextManager.service(ContextManager.java(Compiled
Code))
at
org.apache.tomcat.service.connector.Ajp13ConnectionHandler.processConnection
(Ajp13ConnectionHandler.java:160)
at
org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoint.java:416)
at
org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPool.java:501)
at java.lang.Thread.run(Thread.java:498)

Re: jasper: weird behaviour

[Jeff Kilbride]

You might be interested in this post from Tomcat-dev.

--jeff

---
Hi All!

Different encodings support in Servlet/JSP is an ancient well-known problem.
The setCharacterEncoding() method of HttpServletRequest allows to change
request
encoding before reading parameters. Thus, servlet is able to change encoding
in
accordance with its needs. (Small lyrical digression: what does this
encoding mean?
I'll post my thoughts about it separately)
Howevet the problem still exists in JSP (there were several postings about
the problem in
this maillist). The purpose of this mail is to propose a solution for
encodings support in JSP.

Problem description
===
A JSP programmer is not able to change request encoding for incoming JSP
request, since
"This method [setCharacterEncoding] must be called prior to parsing any post
data or
reading any input from the request. Calling this method once data has been
read will
not affect the encoding." (Servlet 2.3 Spec). This happens because request
parameters
being read inside org.pache.jasper.servlet.JspServlet, before calling
generated JSP-servlet.
As a result we have the following behaviour of compiled JSP for non-English
environments:
1) incoming request being read using 'ISO-8859-1'
2) getParameter() method returns a value in 'ISO-8859-1', but JSP-servlet
suppose the
   return value has JVM default encoding (say "KOI8-R") -- here is ???
instead of
   real parameter value. Here is a problem.

Problem solution

There should be a configurable optional parameter for JspServlet (say
'requestEncoding') to
change request encoding. According to this parameter JspServlet should call
setCharacterEncoding()
before processing request. It does not conflict with JSP 1.2 Spec, since
there are now any
words about default encoding of incoming request over there.

I have made neccessary changes to implement this feature in
tomcat-4.0-20010807. It works fine
with different Cyrillic encodings. (Suppose the same result for the rest of
non-Latin1 encodings).
I clearly understand that proposed solution is not a panacea and it's a
subject to discuss.


Regards,
Andrey Aristarkhov


Diffs are followed (also as attachments). I have also attached a sample JSP
for encoding testing.


file: org/apache/jasper/EmbededServletOptions.java

147a148,152
>  * Java platform encoding for incoming request.
>  */
> private String requestEncoding;
>
> /**
219a225,228
> public String getRequestEncoding() {
> return requestEncoding;
> }
>
320a330
> this.requestEncoding = config.getInitParameter("requestEncoding");

file: org/apache/jasper/EmbededServletOptions.java

144a145,149
>
> /**
>  * Java platform encoding for incoming request.
>  */
> public String getRequestEncoding();

file: org/apache/jasper/servlet/JspServlet.java

422c422,426
< String includeUri
---
> // According to section 4.9 of Servlet 2.3 spec we have to
> // setCharacterEncoding() before reading any parameter
> if (options.getRequestEncoding()!=null)
>   request.setCharacterEncoding(options.getRequestEncoding());
> String includeUri

Re: Fine tuning my Apache -Tomcat Server

[Jeff Kilbride]

Turn off servlet auto-reloading as well.

I would recommend upgrading to 3.2.3, because of security problems and other
bug fixes with 3.2.1 and 3.2.2. Just copy the new .jar files from 3.2.3 to
your 3.2.1 installation and restart. It's that simple. (It's outlined in the
3.2.3 release notes...)

--jeff

- Original Message -
From: "Martin van den Bemt" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 08, 2001 7:13 AM
Subject: RE: Fine tuning my Apache -Tomcat Server


> Which connector are you using (ajp12 or ajp13). The last one should be a
lot
> faster (ajp12 is frozen anyway). Also turning of unecessary debugging /
> logging in this area will speed up the process..
>
> Mvgr,
> Martin
>
> > -Original Message-
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED]]
> > Sent: Wednesday, August 08, 2001 4:12 PM
> > To: [EMAIL PROTECTED]
> > Subject: Fine tuning my Apache -Tomcat Server
> >
> >
> > Hi,
> >
> > I have Apache and Tomact running on Linux server.
> >
> > The following are the details of the same
> >
> > Apache:
> > Version : Apache_1.3.12-i686
> >
> > Tomcat:
> > Version: Jakarta-tomcat-3.2.1
> >
> > Linux:
> > Version :6.2
> >
> > Java:
> > Version : JDK 1.2.2
> >
> > Hardware Configuration:
> > Processor : Intel PIII
> > CPU Cycles : 733 M Hz.
> > RAM:256MB
> > Hard Disk : 256 MB
> >
> > Applications :
> > Technology : Java Servlet's
> >
> > Users:
> > 100 users for every 10 minutes (approx)
> >
> > Network :
> > Leased Line
> > Band Width : 2 MegaBytes/Second
> >
> > My Current Problem:
> > There  are constant complaints that the server is slow in delivering the
> > response to the users. But all the applications are working fine.
> >
> > Please help me out in fine tuning my Apache -Tomcat Server.
> > Any inputs for improving server performance are highly appreciated.
> >
> > Thanks and Regards
> > Srinivas Chebolu
> >
> >
> >
>

Re: PLEASE UNSUBSCRIBE ME

[Jeff Kilbride]

Since we filtered HTML, can we filter any message with 'unsubscribe' in the
subject?

--jeff

- Original Message -
From: "Olivier LAUDREN" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, August 07, 2001 5:11 AM
Subject: RE: PLEASE UNSUBSCRIBE ME


> This could be found in the mail's header:
> list-unsubscribe: 
>
> -Message d'origine-
> De : Jockel, Jeff [mailto:[EMAIL PROTECTED]]
> Envoyé : mardi 7 août 2001 14:11
> À : [EMAIL PROTECTED]
> Objet : PLEASE UNSUBSCRIBE ME
>
>
>
>
> -Original Message-
> From: Beth Kelly [mailto:[EMAIL PROTECTED]]
> Sent: Monday, August 06, 2001 6:19 PM
> To: [EMAIL PROTECTED]
> Subject: Re: JDBC Realms
>
>
> Actually, I see why you would not want the passwords in memory.
>
> Kyle Wayne Kelly
> (504)391-3985
> http://www.cs.uno.edu/~kkelly
> - Original Message -
> From: "Michael Wentzel" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Monday, August 06, 2001 12:13 PM
> Subject: RE: JDBC Realms
>
>
> > > > Advantage: You don't lose existing session data
> > > > Disadv   : You're not actually re-authenticating
> > > (not really authenticating, you lost me)
> >
> > After looking at some code I figured something out...
> > I was thinking about this architecture wrong.  Kyle was
> > right just using:
> >
> > session.setAttribute("j_password", sPassword);
> >
> > will provide a hook for password changes.
> >
> >
> > ---
> > Michael Wentzel
> > Software Developer
> > Software As We Think - http://www.aswethink.com
> >
>

Re: Servllet

[Jeff Kilbride]

FWIW, I had to write a class to invoke a method on a given object at timed
intervals using the Reflection API. I use it to "automatically" refresh bean
info from my database at set intervals. I'm attaching the code, called
TimedMethodInvoker.java, and a simple example.

Works like a charm for me. Hope it helps.

--jeff

- Original Message -
From: "Michael Wentzel" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, August 06, 2001 5:27 AM
Subject: RE: Servllet


> > Maybe do a sleep instead of busy waiting (looping takes up
> > too much cpu
> > time).
> >
> > > > How about running an application that periodically posts to
> > > > the servlet?
> > > >
> > > >
> > > > > How can a servlet be automatically be invoked by
> > > > > itself, say peroidically?
> > > > > Usul
> > >
> > > I think a better way is to write a servlet which is configured
> > > as a load on startup servlet which forks a separate "scheduler"
> > > thread which then defines tasks to be completed at specified
> > > interval(using a properties file or init params...).  The
> > > Thread(Runnable) will essentially be a infinite loop which will
> > > check to see if it's time to run certain tasks. i.e.
> > >
> > > while(1) {
> > > if ( /*somethings true 1*/ ) {
> > > /*perform task 1*/
> > > }
> > > if ( /*somethings true 2*/ ) {
> > > /*perform task 2*/
> > > }
> > > // ...
> > > }
>
> Yeah, that's the general idea.  That was the //... part any additional
> cleanup between states and a wait state.  Otherwise the process, depending
> on the performance of your separate scheduled jobs and the frequency of
> them, could pretty much lock the processor down.  This is meant as a
> rudamentary example.  This is the generic model of a scheduling
> thread/server.
>
>
> ---
> Michael Wentzel
> Software Developer
> Software As We Think - http://www.aswethink.com
>

 TimedMethodInvoker.java
 ExampleBean.java

Re: Upgrade to Tomcat DECREASES Performance?

[Jeff Kilbride]

Good enough to have answered this question several times. Have you checked
the archives?

http://mikal.org/interests/java/tomcat/archive/view?mesg=35371
http://mikal.org/interests/java/tomcat/archive/view?mesg=33145
http://mikal.org/interests/java/tomcat/archive/view?mesg=30821
http://mikal.org/interests/java/tomcat/archive/view?mesg=26859

Other things I would suggest:

-- If you're using ajp13, you can tune down the number of threads for your
ajp12 connector

-- Use the mod_jk module that comes with TC 3.3. It allows you to stop and
start TC without restarting Apache. You can download the 3.3 distribution
and make it yourself, or just use the pre-compiled binary for Linux (which
is what I use). It's compatible with 3.2.x and can be found here:

http://jakarta.apache.org/builds/jakarta-tomcat/release/v3.3-b1/bin/linux/i3
86/

-- Upgrade to 3.2.3. It's really easy. You just copy the jar files into your
existing installation. The release notes explain it.

-- Monitor the tomcat-dev list. You learn a lot more there than you do on
the user list (even though you don't normally post there...)

Thanks,
--jeff

- Original Message -
From: "Dwight Powell" <[EMAIL PROTECTED]>
To: "Tomcat Users" <[EMAIL PROTECTED]>
Sent: Monday, July 30, 2001 10:09 AM
Subject: Upgrade to Tomcat DECREASES Performance?


> OK, let's see how good you folks are:
>
> I work for a large national not-for profit membership organization.  We've
> got a large database application that we serve to our members from a small
> group of servers out of our national office.  About a week ago, we
upgraded
> our servers, moving from Apache JServ/mod_jserv/AJP12 to
> Tomcat/mod_jk/AJP13.  Ever since the upgrade, our servers have been HIGHLY
> unstable (our 3 busiest servers have each crashed at least once a day for
> the last week)  and we're HIGHLY confused.  The only real change has been
> the move to Tomcat, which we thought was a large upgrade, but our app has
> been acting as if we downgraded!
>
> Before the upgrade, application was having some performance issues (record
> locking, result set not found errors), probably caused by the large load
on
> our servers combined with some inefficiencies in the code.  After the
> upgrade, the same performance errors exist, but seem to be magnified, as
if
> Tomcat is working SLOWER than JServ.  But here's the worst part:
>
> At least once a day (usually after running for a while), the application
> will simply stop responding - users will attempt to save forms and/or
> navigate between pages and will never get a respond (eventually they'll
get
> a "Page Cannot Be Displayed Error").  When the server is in this state, we
> check the running processes and we see lots of java threads loaded and
> waiting for response, but none of them is RUNNING, so I don't THINK this
is
> a runaway thread problem.  Once the server hangs like this, our processor
> usage by the JVM will go down to 0% and stay there until we restart Apache
> and Tomcat.
>
> Here's our specs:
>
> SERVER SPECS (average):
>
> Dual Pentiums (1 Ghz)
> 2.5 GB RAM
> Red Hat Linux 7.1
>
> OLD CONFIGURATION:
> Apache 1.3.14
> Apache JServ (with mod_jserv and ajp12)
> IBM JDK 1.3-5
> Cloudscape 3.5
>
> NEW CONFIGURATION:
> Apache 1.3.14
> Jakarta Tomcat 3.2.2 (with mod_jk and ajp13)
> IBM JDK 1.3-9
> Cloudscape 3.5
>
> So... does anyone out there have any idea why we're seeing these problems?
> Where should we focus our investigation?   Should we just go back to
JServ,
> or is Tomcat REALLY better?
>
> Thanks to one and all...
>
> ***
> Dwight Powell
> Manager of Systems / Lead Programmer
> NACCRRA (www.naccrra.org)
> [EMAIL PROTECTED]
>
>

Re: Apache cannot connect to Tomcat

[Jeff Kilbride]

You're right, Andrew! I misread and thought Kelly hadn't added _any_
connectors to server.xml. My mistake.

Kelly, I have JSP's running with 3.2.3 on RedHat & Apache 1.3.19. Can you
post the relevant parts of your server.xml, httpd.conf and web.xml files?

--jeff

- Original Message -
From: "Andrew Robson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, July 29, 2001 12:56 PM
Subject: Re: Apache cannot connect to Tomcat


> Hang on Jeff this isn't correct is it? mod_jk can support either the ajp12
> or ajp13 protocol so adding the ajp13 connector to server.xml isn't likely
> to help.
> I'm afraid I can't help with the jsp side of things as I only ever
> write servlets. Might be interesting to see if you can get a servlet
> to work to see if JSP issue.
>
> andrew
>
> On Sun, 29 Jul 2001, you wrote:
> > If you're using mod_jk, you have to add the connector to your server.xml
> > file. I'm not sure why you *haven't* done that. Apache and Tomcat won't
be
> > able to communicate without it. So, I would suggest starting there.
> >
> > --jeff
> >
> > - Original Message -
> > From: "Kelly E. Grooms" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Saturday, July 28, 2001 2:50 PM
> > Subject: Apache cannot connect to Tomcat
> >
> >
> > > Hello,
> > >
> > > I've scoured the web and all the Tomcat lists that I can find and
still
> > > cannot solve this error.  Everything works fine through port 8080 as
> > > stand-alone, but when I try accessing Tomcat through Apache the page
times
> > > out with no output from the JSP.  Afterward my mod_jk.log file
contains:
> > >
> > > [jk_connect.c (143)]: jk_open_socket, connect() failed errno = 110
> > > [jk_ajp12_worker.c (152)]: In jk_endpoint_t::service, Error sd = -1
> > >
> > > I compiled mod_jk.so myself.  I've added mod_jk.conf-auto to my
httpd.conf
> > > file.  I altered workers.properties to reflect my JAVA_HOME,
TOMCAT_HOME
> > and
> > > file separator ("/").  I have _not_ added the connector for ajp13 to
my
> > > server.xml file.
> > >
> > > Here is my configuration:
> > >
> > > Mandrake Linux 8.0
> > > Apache 1.3.19
> > > Java 1.3.1 (from Sun)
> > > Tomcat 3.2.3
> > >
> > > I've installed older versions of Tomcat on other configurations in the
> > past
> > > and did not have this much trouble.  I'm beginning to feel very
> > discouraged.
> > > Can anyone help?  Thanks.
> > >
> > > Kelly E. Grooms
> > > [EMAIL PROTECTED]
> > >
> --
>
> Andrew Robson
>
> tel: (0141) 424 0607
> mobile: 07759 430234
> email: [EMAIL PROTECTED]
>

Re: HTML in Messages and politeness

[Jeff Kilbride]

> [X]  0 -> I don't give a damn shit.

I do, however, agree with the points on politeness.

I think if you reject HTML-only messages, a lot of new subscribers are not
going to be able to figure out why they can't post to the list. Most
probably don't even realize they are doing it, since the newer email clients
default that way.

"How do I turn off HTML formatting?" might be a good addition to the FAQ and
the auto-generated email that goes out when you sign up, if it's not already
there.

Thanks,
--jeff

Re: Apache cannot connect to Tomcat

[Jeff Kilbride]

If you're using mod_jk, you have to add the connector to your server.xml
file. I'm not sure why you *haven't* done that. Apache and Tomcat won't be
able to communicate without it. So, I would suggest starting there.

--jeff

- Original Message -
From: "Kelly E. Grooms" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, July 28, 2001 2:50 PM
Subject: Apache cannot connect to Tomcat


> Hello,
>
> I've scoured the web and all the Tomcat lists that I can find and still
> cannot solve this error.  Everything works fine through port 8080 as
> stand-alone, but when I try accessing Tomcat through Apache the page times
> out with no output from the JSP.  Afterward my mod_jk.log file contains:
>
> [jk_connect.c (143)]: jk_open_socket, connect() failed errno = 110
> [jk_ajp12_worker.c (152)]: In jk_endpoint_t::service, Error sd = -1
>
> I compiled mod_jk.so myself.  I've added mod_jk.conf-auto to my httpd.conf
> file.  I altered workers.properties to reflect my JAVA_HOME, TOMCAT_HOME
and
> file separator ("/").  I have _not_ added the connector for ajp13 to my
> server.xml file.
>
> Here is my configuration:
>
> Mandrake Linux 8.0
> Apache 1.3.19
> Java 1.3.1 (from Sun)
> Tomcat 3.2.3
>
> I've installed older versions of Tomcat on other configurations in the
past
> and did not have this much trouble.  I'm beginning to feel very
discouraged.
> Can anyone help?  Thanks.
>
> Kelly E. Grooms
> [EMAIL PROTECTED]
>

Re: How to forward with new request/query string?

[Jeff Kilbride]

Hi Erin,

You can't change the original request (at least, I don't think you
can...) -- but you can add info to the request before forwarding it on to
your jsp page. Just add the extra info to the query string of the jsp your
forwarding to:

String location = "myForward.jsp?foo=bar&baz=1";
RequestDispatcher rd = request.getRequestDispatcher(location);
rd.forward(request, response);

The parameters foo and baz will now be available in myForward.jsp, along
with all the original request params. If foo and baz exist in the original
request, calling request.getParameter() will return the newly set value and
calling request.getParameterValues() will return all values for that
parameter.

You can also store info in the request object using setAttribute() and
retrieve them in the jsp by using getAttribute():

(in servlet)
request.setAttribute("key", value);

(in jsp -- if the value is a String object)
String value = (String)request.getAttribute("key");

Notice that you have to cast the object back to the correct type in your
jsp.

Hope this helps.

--jeff

- Original Message -
From: "Erin Lester" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, July 19, 2001 1:36 PM
Subject: How to forward with new request/query string?


> I was wondering how you go about transferring control from a servlet to
> another page (in this case a jsp page) before committing the response.  I
> need to send request variables to this new page.  I've tried using a
> request dispatcher, but I believe that this just sends the request that
> the servlet got and not a new one with the querystring values (which are
> appended to the url I want to forward to) in it.
>
> Can someone tell me how to do this?  Also, what's the difference between
> sendRedirects, dispatcher forwards, etc. ?  and is there any way to
> transfer control to another page and have the user's location bar reflect
> the address of the new page?
>
> Any help would be appreciated!
> - Erin
>

Re: How to change the IP address

[Jeff Kilbride]

Whoops, forgot the closing "/>" for the "inet" param. Should be:



Thanks,
--jeff

- Original Message -----
From: "Jeff Kilbride" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, July 17, 2001 11:33 PM
Subject: Re: How to change the IP address


> The PoolTCPConnector class has a parameter called "inet" that's not really
> documented. I used this parameter to make my ajp12 and ajp13 connectors
> listen only on the localhost interface -- by default, they listen on all
> interfaces, even external, which is bad.
>
> I don't know if it will work, but you could try including the "inet" param
> in the definition of the Normal HTTP connector inside your server.xml
file.
> Maybe like this:
>
> ---
> 
> 
>  value="org.apache.tomcat.service.http.HttpConnectionHandler"/>
>  value="127.0.0.1"
>  value="8080"/>
> 
> ---
>
> Just replace "127.0.0.1" with the IPAddress you want to use.
>
> Thanks,
> --jeff
>
> - Original Message -
> From: "Neelu Shah" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Monday, July 16, 2001 3:51 PM
> Subject: How to change the IP address
>
>
> > Hi All,
> >
> > could someone tell me how to change the default IP address for tomcat
(no
> > apache used)
> > from localhost to another IP address.
> >
> > im running tomcat on Windows NT.
> >
> > thanks
> > N
> >
>

Re: How to change the IP address

[Jeff Kilbride]

The PoolTCPConnector class has a parameter called "inet" that's not really
documented. I used this parameter to make my ajp12 and ajp13 connectors
listen only on the localhost interface -- by default, they listen on all
interfaces, even external, which is bad.

I don't know if it will work, but you could try including the "inet" param
in the definition of the Normal HTTP connector inside your server.xml file.
Maybe like this:

---





---

Just replace "127.0.0.1" with the IPAddress you want to use.

Thanks,
--jeff

- Original Message -
From: "Neelu Shah" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 16, 2001 3:51 PM
Subject: How to change the IP address


> Hi All,
>
> could someone tell me how to change the default IP address for tomcat (no
> apache used)
> from localhost to another IP address.
>
> im running tomcat on Windows NT.
>
> thanks
> N
>

Re: Need workaround for Tomcat security.

[Jeff Kilbride]

Hi Andrew,

I know that there were some security-related problems with 3.2.1 and certain
URLs. I think a bug was found and fixed right around the time of 3.2.2 beta
5. I would suggest upgrading to 3.2.2. It's very painless -- all config
files stay the same, just copy your old ones into your 3.2.2 install
directory and change TOMCAT_HOME. I'm not seeing the problem on my
installation (TC 3.2.2, Linux, apache, mod_jk).

Thanks,
--jeff

- Original Message -
From: "Andrew Robson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 16, 2001 1:39 PM
Subject: Re: Need workaround for Tomcat security.


> Jeff,
>TC 3.2.1 on linux.
>Apache and mod_jk
> It seems to me (without having had a chance to check)
> that this must be a misconfig at the apache
> and apache/tomcat end of things rather than a tomcat bug as such.
>
> Any thoughts? It would be a pretty big hole if it was a genuine
> bug.
>
> andrew
>
> On Mon, 16 Jul 2001, you wrote:
> > Andrew,
> >
> > What version of Tomcat did this affect Form-based authentication on? I
tried
> > the URL patterns mentioned on my Form-based Realm, and the Realm worked
> > correctly -- no security problems. I'm using TC 3.2.2 on Linux.
> >
> > Thanks,
> > --jeff
> >
> > - Original Message -
> > From: "Andrew Robson" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Monday, July 16, 2001 7:29 AM
> > Subject: Re: Need workaround for Tomcat security.
> >
> >
> > > Hi,
> > >   No workaround I'm afraid. I can confirm that the problem
> > > affects form - based JDBCRealm as well. Tried putting
> > > */admin/* into url pattern and broke security completely.
> > > I wonder whether a JkMount directive with approriately
> > > placed wildcards might work but haven't had time to try.
> > > I'd be very interested if you find a solution.
> > > Presumably no-one on the list has one?
> > >
> > > andrew
> > >
> > > On Sun, 15 Jul 2001, you wrote:
> > > > Ok, i needed to put some security constraints to a dircetory, so I
added
> > this
> > > > to my web.xml:
> > > >  
> > > >   UQoS Amin Area
> > > >   
> > > >  UQoS Amin Area
> > > >/admin/*
> > > >   
> > > > I use BASIC authentication using the memory realm.
> > > > Works like it supposed to when someone goes to my
> > http://xxx/webapp/Admin/ or
> > > > something below, HOWEVER, if they type http://xxx/webapp//Admin/ (or
> > even
> > > > more slashes), all security checkings are bypassed, anyone arr let
right
> > in !
> > > > (same things happens always, try it with the 'security' example
shipped
> > with
> > > > Tomcat.
> > > > Sever bug!, I have posted it to BugZilla. This applies to atleast
Tomcat
> > > > 3.2.1 and 3.2.2.
> > > > And I need it fixedas soon as possible. Does anyone know a
workaround to
> > > > thisone.(I'd rather not upgrade to Tomcat 4 yet,seems like its fixed
> > here.)
> > > > --
> > > > Nils O. Selåsdal
> > > --
> > >
> > > Andrew Robson
> > >
> > >
> > >
> --
>
>
>

Re: ajp13 dies unexpectately

[Jeff Kilbride]

How many threads have you specified as max_threads in your PoolTCPConnector
for ajp13 in your server.xml file? If you haven't done this, and you're
running a reasonably high load, you should take a look at the minimal TC
user's guide here:

http://jakarta.apache.org/tomcat/tomcat-3.2-doc/uguide/tomcat_ug.html

The last example at the bottom of the page shows how to configure
max_threads, max_spare_threads, and min_spare_threads for the
PoolTCPConnector class. Tomcat doesn't react gracefully if the max_threads
number is ever reached and exceeded. I believe the default max is 50
threads. If you're getting more than 50 concurrent connections, you need to
raise this number or you'll have problems.

Thanks,
--jeff

- Original Message -
From: "maarten hartsuijker"
<[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 16, 2001 2:00 AM
Subject: ajp13 dies unexpectately


> we have been running tomcat 3.2.2 with apache, mod_jk and ajp13 support
for
> about 3 weeks now. In those weeks it has died 2 times unexpectately and
the
> only thing the mod_jk.og is showing me is a jk_ajp13_worker.c (586)]:
Error
> connecting to the Tomcat process.
>
> With a netstat I can see that 8009 is not up anymore, so both times I have
> issued a shutdown and a start. That works but (ofcourse) I'd rather not
have
> this problem at all.
>
> Anyone had this before and, better yet, found a solution?
>
> kind regards,
>
> Maarten Hartsuijker
>

Re: Need workaround for Tomcat security.

[Jeff Kilbride]

Andrew,

What version of Tomcat did this affect Form-based authentication on? I tried
the URL patterns mentioned on my Form-based Realm, and the Realm worked
correctly -- no security problems. I'm using TC 3.2.2 on Linux.

Thanks,
--jeff

- Original Message -
From: "Andrew Robson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 16, 2001 7:29 AM
Subject: Re: Need workaround for Tomcat security.


> Hi,
>   No workaround I'm afraid. I can confirm that the problem
> affects form - based JDBCRealm as well. Tried putting
> */admin/* into url pattern and broke security completely.
> I wonder whether a JkMount directive with approriately
> placed wildcards might work but haven't had time to try.
> I'd be very interested if you find a solution.
> Presumably no-one on the list has one?
>
> andrew
>
> On Sun, 15 Jul 2001, you wrote:
> > Ok, i needed to put some security constraints to a dircetory, so I added
this
> > to my web.xml:
> >  
> >   UQoS Amin Area
> >   
> >  UQoS Amin Area
> >/admin/*
> >   
> > I use BASIC authentication using the memory realm.
> > Works like it supposed to when someone goes to my
http://xxx/webapp/Admin/ or
> > something below, HOWEVER, if they type http://xxx/webapp//Admin/ (or
even
> > more slashes), all security checkings are bypassed, anyone arr let right
in !
> > (same things happens always, try it with the 'security' example shipped
with
> > Tomcat.
> > Sever bug!, I have posted it to BugZilla. This applies to atleast Tomcat
> > 3.2.1 and 3.2.2.
> > And I need it fixedas soon as possible. Does anyone know a workaround to
> > thisone.(I'd rather not upgrade to Tomcat 4 yet,seems like its fixed
here.)
> > --
> > Nils O. Selåsdal
> --
>
> Andrew Robson
>
>
>

Re: Okay. this is getting annoying.

[Jeff Kilbride]

>From your previous posts, you're server.xml entry is not correct. You can't
reference multiple IP's, domains, etc.. in the  directive.

-
My most recent attempt had this in server.xml



  
-

If you want to access this as "www.pedsforless.com" and "pedsforless.com",
you need to set up two independent  directives. (I'm assuming you're
using TC3.2.2)  Also realize that this will instantiate all of your servlets
twice under two different contexts.

Go to the archives here:

http://mikal.org/interests/java/tomcat/index.jsp

and search on "virtual host jeff". This will pull up all my previous posts
on virtual hosting, including examples of my config.

Thanks,
--jeff

- Original Message -
From: "Chuck Cochems" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, July 15, 2001 2:09 PM
Subject: Okay. this is getting annoying.


> I've posted the snippets from my conf files, and still haven't gotten an
> answer to why i cannot access a virtual host's servlets and JSPs except
> by IP address.  I haven't seen the problem discussed in the archives
> either.  I know this mailing list is pretty busy, but i don't have time
> to wait around forever.
>
> To sum up. I f I put the virtual hosts IP address in a context, I can
> access JSP via the IP address, whether or not the virtualhost directive
> has the IP address, or the domain name.  If I only put a context for the
> domain name, I cannot access it via IP address OR domain name.  If I put
> contexts for both, in anu order, the IP address works, the domain does
> not.
>
> My syntax has all been lifted straight form the samples in the docs on
> the site.  What could possibly be wrong?
>

Re: help..

[Jeff Kilbride]

The MySQL JDBC driver is a type 4 driver, so it will work on Windows and
Linux. Just put the MySQL jar file in your classpath on your Windows
machine.

Thanks,
--jeff

- Original Message -
From: "Ben Kimball" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, July 15, 2001 11:24 AM
Subject: Re: help..


> You will need to install the MySQL JDBC driver on every box that will be
> connecting to the SQL server. Go to the mysql web site and look for the (I
> assume wondog means windows?) Windows driver. Or in some situations you
can
> use a Type 4 JDBC driver which can be downloaded by your users from your
web
> server, but probably easier to use the dedicated Windows JDBC driver (aka
a
> Type 2 JDBC driver) to get started and make sure everything works.
>
> On Sunday 15 July 2001 10:51, you wrote:
> > Hi
> > i have a linux-box with java+tomcat+jdbcand working fine...
> > question : how can i connect from other computer system to my linux box
> > through jdbc?
> > Do i need to install another driver to the other one?
> > i was trying to connect from my windog98se to my server... so.. compiled
> > some simple connection java code on wondog box, then i have an error
> > message "org.gjt.mm.mysql.Driver No suitable driver"...
> >
> > How can i make them connect each other other?
> >
> > Thanks in advance..
>

Re: java-linux-tomcat configuration problem

[Jeff Kilbride]

I would try the IBM JDK before Blackdown (before Sun, too...), but that's
just my personal choice.

The Blackdown port can be found at www.blackdown.org.

--jeff

- Original Message -
From: "Adam Fowler" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, July 14, 2001 3:47 AM
Subject: Re: java-linux-tomcat configuration problem


> Hi,
>
> Many ppl using 1.3.1 specifically have noted huge memory usage increases
> (leaks) which either take a long time to clean themselves up or are
permanent.
>
> The issue has been reported by ppl using JNI, tomcat, app contexts etc etc
so
> is a JDK bug.
>
> Sun still haven't fixed it (I just checked) Maybe 1.4 fixes it, but that
is
> still in Beta so I wouldn't use it on a deployed platform just yet.
>
> Apparently using 1.3 fixes it, but its a real pain to find. Sun don't
> actually advertise its whereabouts on their site.
>
> The blackdown JDK doesn't seem to have this problem so you could try that
(I
> don't know where it can be found)
>
> Hope that helps,
>
> Adam.
>
> 
> Adam Fowler
> Help Desk Live Project
> Information Services
> University of Wales, Aberystwyth
> Web guy+author on the TomcatBook Project
> http://tomcatbook.sourceforge.net
> e-mail: [EMAIL PROTECTED]
> 
>
> On Friday 13 July 2001 16:33, you wrote:
> > What is the infamous bug for jdk?
> >
> > Thanks!
> > Dan
> >
> >
> > From: Adam Fowler <[EMAIL PROTECTED]>
> >
> > >Reply-To: [EMAIL PROTECTED]
> > >To: [EMAIL PROTECTED]
> > >Subject: Re: java-linux-tomcat configuration problem
> > >Date: Thu, 12 Jul 2001 22:09:53 +
> > >
> > >Hi,
> > >
> > >You have two versions of httpd running!?! Make sure that if you have
two
> > >version using tomcat then you have two tomcat's running. That might
answer
> > >it.
> > >
> > >Otherwise it might be the infamous bug to do with the 1.3.1 JDK
(although,
> > >again, I had no problems on Mandrake! Must be the pretty penguins) You
> > > ould try the 1.3 JDK (if u can find it!) or *cringe* the 1.4 JDK -
> > > although not supported.
> > >
> > >Adam.
> > >
> > >On Thursday 12 July 2001 13:01, you wrote:
> > > > Hello again:
> > > >
> > > > Quick thanks for your responses  .. now more background, as
requested.
> > > >
> > > > a) The contexts that I am running are the ones that came with
apache,
> > > > outlined in server.xml.  I was able to start up those instances with
> > > > jre (but not the example jsps, another problem).
> > > >
> > > > b) I am running Red Hat 6.2.
> > > >
> > > > c) I am using Sun's jdk 1.3.1
> > > >
> > > > d)  Adam, here is the top and ps -x output (the java threads are
> > > > towards the end in ps -x, and are at the top of "top"):
> > > >   PID TTY  STAT   TIME COMMAND
> > > > 1 ?S  0:02 init
> > > > 2 ?SW 0:00 [kflushd]
> > > > 3 ?SW 0:01 [kupdate]
> > > > 4 ?SW 0:00 [kpiod]
> > > > 5 ?SW 0:00 [kswapd]
> > > > 6 ?SW<0:00 [mdrecoveryd]
> > > > 7 ?SW<0:00 [raid1d]
> > > > 8 ?SW<0:00 [raid1d]
> > > > 9 ?SW<0:00 [raid1d]
> > > >10 ?SW<0:00 [raid1d]
> > > >   171 ?S  0:00 syslogd -m 0
> > > >   180 ?S  0:00 klogd
> > > >   209 ?S  0:00 /usr/sausalito/sbin/cced
> > > >   586 ?S  0:00 crond
> > > >   598 ?S  0:00 inetd
> > > >   628 ?S  0:00 named
> > > >   644 ?S  0:00 /usr/sbin/dhcpd -q eth0
> > > >   649 ?S  0:01 /usr/sbin/ahttpd -f
> > >
> > >/etc/admserv/conf/httpd.conf
> > >
> > > >   673 ?S  0:01 /usr/sbin/httpd -f
> > > > /etc/httpd/conf/httpd.conf 705 ?S  0:00 sendmail:
accepting
> > > > connections
> > > >   717 ?S  0:00 sh /usr/bin/safe_mysqld
> > >
> > >--datadir=/var/lib/mysql
> > >
> > > > --pi
> > > >   784 ?S  0:00 /usr/sbin/atalkd
> > > >   807 ?S  0:00 smbd -D
> > > >   816 ?S  0:00 nmbd -D
> > > >   825 ?S  0:00 /sbin/lcdsleep
> > > >   867 ?S  0:00 /sbin/consoled /sbin/getty ttyS0 115200
> > > >   904 ?S  0:00 /usr/sbin/afpd -U
uams_clrtxt.so,uams_dhx.so
> > >
> > >-g
> > >
> > > > guest
> > > > 1353 ?S  0:00 in.telnetd: 10.6.18.30
> > > > 1354 pts/0S  0:00 login -- admin
> > > > 1758 ?S  0:00 in.telnetd: 10.6.18.30
> > > > 1759 pts/1S  0:00 login -- admin
> > > > 1793 pts/1S  0:00 su
> > > > 1794 pts/1S  0:00 bash
> > > > 1952 ?S  0:00 smbd -D
> > > > 2283 pts/0S  0:00 su
> > > > 2284 pts/0S  0:00 bash
> > > > 2403 pts/0R  2:07
> > >
> > >/usr/java/jdk1.3.1/bin/i386/native_threads/java
> > >
> > > > -Dtom
> > > > 2448 pts/0S  0:00
> > >
> > >/usr/java/jdk1.3.1/bin/i386/native_threads/java
> > >
> > > > -Dtom
> > > > 2449 pts/0S  0:00
> > >
> > >/usr/java/jdk1.3.1/bin/i386/native_threads/java
> > >
> > > > -Dtom
> > > > 2450 pts/0S   

Re: apache + tomcat + virtual hosts

[Jeff Kilbride]

Hi Connie,

I think the first thing you should do is read the Apache docs on virtual
hosting here:

http://httpd.apache.org/docs/vhosts/index.html

Then go to the archive here:

http://mikal.org/interests/java/tomcat/index.jsp

and do a search on "virtual host jeff" -- this will pull up all my previous
posts on virtual hosting, including examples of my config.

Thanks,
--jeff

- Original Message -
From: "Connie Chan" <[EMAIL PROTECTED]>
To: "Jack Hui" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Friday, July 13, 2001 6:04 AM
Subject: RE: apache + tomcat + virtual hosts


> Jack,
>
> I try a few configuration settings:
>
> Configuration 1:
> I put the default server name in the NameVirtualHost and then create the
> virutal host tag for the virtual host 
>  .  It does not work.  When I browse the default web
> server, the virtual host displays.
>
> Configuration 2:
> I put the default server name in the NameVirtualHost and then create the
> virutal host tag for the default server 
>   and the virtual host 
>  .  It does not work.  When I browse the default web
> server, I get 403 Forbidden error (The error displays You don't have
> permission to access / on this server).
>
> Configuration 3:
> I put '*' in the NameVirtualHost and then create the virutal host tags
> for the default server    and the
> virtual host   .  It does not work.
> Both servers give me 403 Forbidden error.
>
>
> Is the configuration 2 correct?  Should I create another server name as
> the default server name and then have different name for my default web
> server?  Please help.
>
> Thanks,
> Connie
>
>
> -Original Message-
> From: Jack Hui [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, July 12, 2001 9:32 PM
> To: '[EMAIL PROTECTED]'; Connie Chan
> Subject: RE: apache + tomcat + virtual hosts
>
>
> Connie,
>
> Your default server and virtual server are using the SAME IP AND SAME
> PORT
> no., right ?
> What did you put in the NameVirtualHost ? should be the IP of the
> default
> server, right ?
>
> According to the documentation, you have to set the IP to either way (
> not
> 100% sure, but I read before )
>
> But, why don't you put your default sever into the virtual host too ??
> If it
> can be only recognize by the NAME
> the client enter in the browser.
>
> Another solution is you are going to set another IP for all virtual
> host,
> but it involves modifying the DNS entries.
>
> Jack
>
> -Original Message-
> From: Connie Chan [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, July 12, 2001 9:25 PM
> To: [EMAIL PROTECTED]
> Subject: apache + tomcat + virtual hosts
>
>
> Hi,
> Currently, I'm using apache with tomcat.  I have set up the server such
> that it serves a default web server, ssl server (being set up as a
> virtual host with using port 443), and a virtual host (with using port
> 80).  But the document root for default web server and virtual host is
> mixed up. My virtual host and default web server are using the same IP
> address. When I type the default web server, the welcome file for
> virtual host displays. When I type the virtual host, the welcome file
> for the virtual host displays as well. However, if I specify port 8080
> (the HTTP port for tomcat) in the url (for testing tomcat only without
> going thru apache), the default web url would display the correct
> welcome page.
> In my httpd.conf, I have declared
> DocumentRoot d:/jakarta-tomcat/webapps/myapp
> AND
> 
> ServerName vh1.mycomp.com
> DocumentRoot d:/jakarta-tomcat/webapps/vh1
> JkMount /*.jsp ajp12
> JkMount /servlet/* ajp12
> JkMount /email/* ajp12
> 
> AllowOverride None
> deny from all
> 
> 
> AllowOverride None
> deny from all
> 
> 
>
> Do I miss anything?
>
> Thanks,
> Connie
>

Fw: mod_jk.so-eapi vs mod_jk.so-noeapi

[Jeff Kilbride]

Hi Vinay,

EAPI means "extended API". You'll need to use mod_jk.so-eapi if you are
using Apache with mod_ssl for SSL (https) support. If you are not using
mod_ssl, you can use the mod_jk.so-noeapi.

Thanks,
--jeff

> - Original Message -
> From: "Vinay Menon" <[EMAIL PROTECTED]>
> To: "Tomcat Dev" <[EMAIL PROTECTED]>; "Tomcat User"
> <[EMAIL PROTECTED]>
> Sent: Monday, July 09, 2001 2:27 PM
> Subject: mod_jk.so-eapi vs mod_jk.so-noeapi
>
>
> > What exactly are these 2 so's? Why 2 of them and which one should one
use?
> >
> > Would appreciate if somebody could let me know!
> >
> > Thanks
> >
> > Vinay
> >
>

Re: AW: List traffic et al

[Jeff Kilbride]

In a perfect world, yes, I totally agree. However, in the real world, who's
going to enforce the guidelines? (we already have guidelines that few people
pay attention to...) Reposting the guidelines and links to the FAQ and
archive on a regular basis would probably help, and I agree that it should
be done.

Personally, I don't care about the volume on the list. I have a broadband
connection and it only takes me a couple of seconds to download all the
daily messages. However, I think Tomcat is a broad enough subject to warrant
more than just one generic list. I think a couple of focused lists would
help people with non-generic questions. And, as Milt Epstein points out in a
later post, involvement may actually increase on smaller, focused lists.

I don't think the volume on the tomcat-user list will decrease much at all,
even if it is split. However, splitting will allow people with focused
interests beyond getting Tomcat up and running to participate without being
deluged with messages they may not be interested in.

More choices are usually better than less.

--jeff


> From: "Nico Wieland" <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Fri, 6 Jul 2001 13:17:13 +0200
> To: <[EMAIL PROTECTED]>
> Subject: AW: List traffic et al
> 
> i agree 100%. i think a _good_ thing would be to include these guidelines in
> the confirmation message one receives after subscribing to the list. this is
> the way how it's done eg. on sun-managers, they post the guidelines once a
> month. it's the most disciplined list i know.
> 
> -nico
> 
>> Rather let all of us try to reduce volume(both in size and
>> number), it does
>> not require much effort.
>> 
>> Lets just try to follow the following.
> 
> [snip]
> 

Re: List traffic et al

[Jeff Kilbride]

I'm not sure that I agree with the idea that I have to wade through all
these messages "for the good of the list". As I said, I have no experience
with Tomcat on Windows, so I'm not interested in Windows specific issues nor
can I help to solve them. Splitting along platform lines should retain a
relatively good mix of newbies and experienced developers in each list, so I
don't see the problem you are pointing out.

In any event, even if the list is split on basic vs. advanced topics,
advanced users who wanted to help the community could subscribe to all
lists. I don't think it's necessarily "bad" to want to subscribe only to the
lists you feel you can learn from. I've been lurking and posting for about 9
months now and it seems that the same basic group of people answer a
majority of the questions. I used to answer a lot more than I do now, but
I'll admit that I get pretty frustrated answering the same questions over
and over when I know the answers can be easily found in the archive. Is it
selfish or bad of me to skip over questions I used to answer, or should more
responsibility be placed on the person asking the question? (rhetorical,
because there's no way to control whether or not people actually search the
archives before posting...)

A digest version is already available for this list.

--jeff

> From: Dmitri Colebatch <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Fri, 6 Jul 2001 16:03:14 +1000
> To: [EMAIL PROTECTED], Jeff Kilbride <[EMAIL PROTECTED]>
> Subject: Re: List traffic et al
> 
> I think the big problem with splitting the list is that everyone is going to
> be interested in their own little niche.  I for instance learn nothing by
> answering many questions that I answer, but I do learn things from reading
> other answers.  If the list was split, I would (potentially) have the option
> to only see the messages that I would learn from.  This disadvantages other
> people on the list.  And this is going to occur at all levels.  Even relative
> newbies should be capable of answering some questions that they have just
> dealt with the day before.
> 
> If we want to reduce traffic surely a digest is the option.  The JBoss list
> has a _lot_ more traffic than this one, and I am able to receive that in
> digest mode quite happily.
> 
> anyway, my 2c
> 
> cheesr
> dim
> 
> 
> On Fri,  6 Jul 2001 15:45, Jeff Kilbride wrote:
>> Even if the list is not split into these specific sub-topics, I would
>> certainly like to see it split along Windows/Unix lines. I use Unix
>> exclusively and I skip over 99% of the Windows questions, because I don't
>> have any experience with Tomcat on that platform. I'm sure Windows users
>> feel the same way about Unix related questions.
>> 
>> Thanks,
>> --jeff
>> 
>>> From: "Hemant Singh" <[EMAIL PROTECTED]>
>>> Organization: Supportscape Inc.
>>> Reply-To: [EMAIL PROTECTED]
>>> Date: Wed, 4 Jul 2001 13:39:13 +0530
>>> To: <[EMAIL PROTECTED]>
>>> Subject: Re: List traffic et al
>>> 
>>> HI:
>>> Heartly agree with yur idea and before this also i keep on getting agree
>>> with same kind of ideas but i just dont know who is the moderator of this
>>> group and how this can be acheived.
>>> Regards
>>> Hemant
>>> - Original Message -
>>> From: "Milt Epstein" <[EMAIL PROTECTED]>
>>> To: <[EMAIL PROTECTED]>
>>> Sent: Tuesday, July 03, 2001 10:41 PM
>>> Subject: Re: List traffic et al
>>> 
>>>> On Tue, 3 Jul 2001, Sam Newman wrote:
>>>>> Given the huge amount of traffic this list generates, I can rarely
>>>>> get involved with the discussions that take place. It occurs to me
>>>>> that there sems to be three major discussion themes on the list as a
>>>>> whole:
>>>>> 
>>>>> 1.) General servlet/jsp development issues and how tomcat affects them
>>>>> 2.) General tomcat configuration issues
>>>>> 3.) Webserver integration issues
>>>>> 
>>>>> I guess as documentation improves (e.g. tomcat book, work by people
>>>>> like Mike Slinn) points 2&3 will become less of an issue. I'm just
>>>>> wondering if there is any millage in perhaps splitting the list into
>>>>> 2 or 3 lists?  Personally, I've got no issues with getting tomcat up
>>>>> and running and so don't care too much about that end of things,
>>>>> however the servlet/jsp development issues is more interesting to
>>>>>

Re: List traffic et al

[Jeff Kilbride]

Even if the list is not split into these specific sub-topics, I would
certainly like to see it split along Windows/Unix lines. I use Unix
exclusively and I skip over 99% of the Windows questions, because I don't
have any experience with Tomcat on that platform. I'm sure Windows users
feel the same way about Unix related questions.

Thanks,
--jeff


> From: "Hemant Singh" <[EMAIL PROTECTED]>
> Organization: Supportscape Inc.
> Reply-To: [EMAIL PROTECTED]
> Date: Wed, 4 Jul 2001 13:39:13 +0530
> To: <[EMAIL PROTECTED]>
> Subject: Re: List traffic et al
> 
> HI:
> Heartly agree with yur idea and before this also i keep on getting agree
> with same kind of ideas but i just dont know who is the moderator of this
> group and how this can be acheived.
> Regards
> Hemant
> - Original Message -
> From: "Milt Epstein" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, July 03, 2001 10:41 PM
> Subject: Re: List traffic et al
> 
> 
>> On Tue, 3 Jul 2001, Sam Newman wrote:
>> 
>>> Given the huge amount of traffic this list generates, I can rarely
>>> get involved with the discussions that take place. It occurs to me
>>> that there sems to be three major discussion themes on the list as a
>>> whole:
>>> 
>>> 1.) General servlet/jsp development issues and how tomcat affects them
>>> 2.) General tomcat configuration issues
>>> 3.) Webserver integration issues
>>> 
>>> I guess as documentation improves (e.g. tomcat book, work by people
>>> like Mike Slinn) points 2&3 will become less of an issue. I'm just
>>> wondering if there is any millage in perhaps splitting the list into
>>> 2 or 3 lists?  Personally, I've got no issues with getting tomcat up
>>> and running and so don't care too much about that end of things,
>>> however the servlet/jsp development issues is more interesting to
>>> me.
>>> 
>>> I don't have too strong an opinion on it, its just that I worry I'm
>>> missing some interesting topics because I don't have the time to
>>> work though all the posts
>> 
>> This idea has come up before, and I think it's one of the best for
>> dealing with the high volume on this list (I guess it's one of the two
>> or three highest volume apache lists).  I even volunteered to take the
>> lead in doing this.  So I sent a note to the list owner explaining the
>> idea.  Unfortunately, I never heard anything back.  Without the list
>> owner's cooperation/participation (or someone who can modify the
>> apache/jakarta mailing lists), it won't be possible to do this.  So,
>> we could do some work on this (i.e. figuring out what separate lists
>> to have), but unless we know that it's going to come to something, it
>> doesn't make sense to do too much work on it.
>> 
>> Milt Epstein
>> Research Programmer
>> Software/Systems Development Group
>> Computing and Communications Services Office (CCSO)
>> University of Illinois at Urbana-Champaign (UIUC)
>> [EMAIL PROTECTED]
> 

Re: tomcat 3.2.2 session already invalidated-serious problem

[Jeff Kilbride]

Why not wrap your call to invalidate() in a try-catch clause? If the
exception occurs, the session is already invalidated -- which is what you're
trying to accomplish anyway.

try {
oldSession.invalidate();
}
catch ( IllegalStateException ise ) { ; }

--jeff

- Original Message -
From: "Its Me.. Karthik" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, June 26, 2001 3:15 AM
Subject: tomcat 3.2.2 session already invalidated-serious problem


>
>
> Hi,
>
>  im usuing tomcat 3.2.1..when i invalidate my session im getting error..
>
> this is my piece of code
>
> if(monitor.containsKey(uname)){
>
>   HttpSession oldSession = (HttpSession)monitor.get(uname);
>
>   oldSession.invalidate();
>
> }
>
>
> the error is
>
>
>
> Internal Servlet Error:
>
>
> javax.servlet.ServletException: setAttribute: Session already invalidated
> at
org.apache.jasper.runtime.PageContextImpl.handlePageException(PageContextImp
l.java:459)
> at
karthik._0002fkarthik_0002fpage_00031_0002ejsppage1_jsp_2._jspService(_0002f
karthik_0002fpage_00031_0002ejsppage1_jsp_2.java:129)
> at org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:119)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.jasper.servlet.JspServlet$JspCountedServlet.service(JspServlet.ja
va:130)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.jasper.servlet.JspServlet$JspServletWrapper.service(JspServlet.ja
va:282)
> at
org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:429)
> at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:500)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.tomcat.core.ServletWrapper.doService(ServletWrapper.java:405)
> at org.apache.tomcat.core.Handler.service(Handler.java:287)
> at org.apache.tomcat.core.ServletWrapper.service(ServletWrapper.java:372)
> at
org.apache.tomcat.core.ContextManager.internalService(ContextManager.java:79
7)
> at org.apache.tomcat.core.ContextManager.service(ContextManager.java:743)
> at
org.apache.tomcat.service.http.HttpConnectionHandler.processConnection(HttpC
onnectionHandler.java:213)
> at
org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoint.java:416)
> at
org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPool.java:501)
> at java.lang.Thread.run(Thread.java:484)
>
> Root cause:
>
> java.lang.IllegalStateException: setAttribute: Session already invalidated
> at
org.apache.tomcat.session.StandardSession.setAttribute(StandardSession.java:
721)
> at
karthik._0002fkarthik_0002fpage_00031_0002ejsppage1_jsp_2._jspService(_0002f
karthik_0002fpage_00031_0002ejsppage1_jsp_2.java:111)
> at org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:119)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.jasper.servlet.JspServlet$JspCountedServlet.service(JspServlet.ja
va:130)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.jasper.servlet.JspServlet$JspServletWrapper.service(JspServlet.ja
va:282)
> at
org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:429)
> at org.apache.jasper.servlet.JspServlet.service(JspServlet.java:500)
> at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
> at
org.apache.tomcat.core.ServletWrapper.doService(ServletWrapper.java:405)
> at org.apache.tomcat.core.Handler.service(Handler.java:287)
> at org.apache.tomcat.core.ServletWrapper.service(ServletWrapper.java:372)
> at
org.apache.tomcat.core.ContextManager.internalService(ContextManager.java:79
7)
> at org.apache.tomcat.core.ContextManager.service(ContextManager.java:743)
> at
org.apache.tomcat.service.http.HttpConnectionHandler.processConnection(HttpC
onnectionHandler.java:213)
> at
org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoint.java:416)
> at
org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPool.java:501)
> at java.lang.Thread.run(Thread.java:484)
>
>
> please help me..very urgent
> karthik
>
>
> --
--
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
>
>
>

Re: limiting instances of java

[Jeff Kilbride]

max_threads
max_spare_threads
min_spare_threads

They're all parameters of the PoolTCPConnector class. If you're trying to
configure the number of threads Tomcat is using, you should probably read
about all three of them. They're all listed in the same section of the
User's Guide.

If you only configure max_spare_threads, you're not doing anything to limit
the total number of threads Tomcat can spawn. You're just adjusting the
number of idle threads it keeps in it's pool waiting for requests.

--jeff

- Original Message -
From: "Charles Williams (CEO)"
<[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, June 24, 2001 11:05 PM
Subject: Re: limiting instances of java


> Actually, I believe that it's max_spare_threads.
>
> chuck
>
> - Original Message -
> From: "Jeff Kilbride" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Sunday, June 24, 2001 8:53 PM
> Subject: Re: limiting instances of java
>
>
> > Try the manual. (do a find on "max_threads")
> >
> > http://jakarta.apache.org/tomcat/tomcat-3.2-doc/uguide/tomcat_ug.html
> >
> > --jeff
> >
> > - Original Message -
> > From: "Dino Ming" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Sunday, June 24, 2001 8:53 AM
> > Subject: Re: limiting instances of java
> >
> >
> > > I have over 35+ instances...too..
> > > Yes, how can we reduce the number of java instances ?
> > >
> > > - Original Message -
> > > From: "Charles Williams (CEO)"
> > <[EMAIL PROTECTED]>
> > > To: <[EMAIL PROTECTED]>
> > > Sent: Sunday, June 24, 2001 11:29 PM
> > > Subject: limiting instances of java
> > >
> > >
> > > > hey,
> > > >
> > > > I just noticed that there are over 20 instances of java running when
i
> > do a
> > > > ps call.  How can I cut that down?
> > > >
> > > > chuck
> > > >
> > > >
> > > >
> > > >
> > >
> >
>

Re: limiting instances of java

[Jeff Kilbride]

Try the manual. (do a find on "max_threads")

http://jakarta.apache.org/tomcat/tomcat-3.2-doc/uguide/tomcat_ug.html

--jeff

- Original Message -
From: "Dino Ming" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, June 24, 2001 8:53 AM
Subject: Re: limiting instances of java


> I have over 35+ instances...too..
> Yes, how can we reduce the number of java instances ?
>
> - Original Message -
> From: "Charles Williams (CEO)"
<[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Sunday, June 24, 2001 11:29 PM
> Subject: limiting instances of java
>
>
> > hey,
> >
> > I just noticed that there are over 20 instances of java running when i
do a
> > ps call.  How can I cut that down?
> >
> > chuck
> >
> >
> >
> >
>

Re: how to determine if tomcat is running

[Jeff Kilbride]

Have you tried 'man ps' on the systems in question to see what the options
should be?

--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: "Thomas Fischer" <[EMAIL PROTECTED]>
Sent: Sunday, June 24, 2001 5:10 AM
Subject: Re: how to determine if tomcat is running


> thanks for your response, but this method is not working on all unix
> systems because some of them only show .../java as result of the ps
> command (even with the options you mentioned).
>
> does anybody know anything else how i can be sure if tomcat is running
> or not???
>
> thanks,
>
> thomas.
>
> Am Samstag, 23. Juni 2001 um 20:17 schrieb Jeff Kilbride:
>
> > You should have a java process running
> > 'org.apache.tomcat.startup.Tomcat'
> > when Tomcat is running. For example, when I run the 'ps' command I see
> > the
> > following:
> >
> > /usr/local/java/IBMJava2-13/jre/bin/exe/java -Xms64M -Xmx128M
> >-Djava.protocol.handler.pkgs=com.sun.net.ssl.internal.www.protocol
> >-Dtomcat.home=/usr/local/java/jakarta-tomcat-3.2.2
> >org.apache.tomcat.startup.Tomcat
> >
> > This is all on one line, of course, and I have to use 'ps awx
> > --cols=250'
> > (Linux) in order to see the entire command line.
> >
> > Hope this helps.
> >
> > Thanks,
> > --jeff
> >
> > - Original Message -
> > From: "Thomas Fischer" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Saturday, June 23, 2001 10:34 AM
> > Subject: how to determine if tomcat is running
> >
> >
> >> i'm writing a script (on a unix-system) which should do different tasks
> >> wether tomcat is running or not. to do so the script has to figure out
> >> the status of tomcat. most daemons use .pid-files or anything similar.
> >> but i found nothing for tomcat.
> >>
> >> i've looked all through the documentation but found nothing about the
> >> runtime status of tomcat. how can i find out if tomcat is already
> >> running?
> >>
> >> thanks for any responses,
> >>
> >> thomas.
> >>
> >
>

Re: how to determine if tomcat is running

[Jeff Kilbride]

You should have a java process running 'org.apache.tomcat.startup.Tomcat'
when Tomcat is running. For example, when I run the 'ps' command I see the
following:

/usr/local/java/IBMJava2-13/jre/bin/exe/java -Xms64M -Xmx128M
   -Djava.protocol.handler.pkgs=com.sun.net.ssl.internal.www.protocol
   -Dtomcat.home=/usr/local/java/jakarta-tomcat-3.2.2
   org.apache.tomcat.startup.Tomcat

This is all on one line, of course, and I have to use 'ps awx --cols=250'
(Linux) in order to see the entire command line.

Hope this helps.

Thanks,
--jeff

- Original Message -
From: "Thomas Fischer" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, June 23, 2001 10:34 AM
Subject: how to determine if tomcat is running


> i'm writing a script (on a unix-system) which should do different tasks
> wether tomcat is running or not. to do so the script has to figure out
> the status of tomcat. most daemons use .pid-files or anything similar.
> but i found nothing for tomcat.
>
> i've looked all through the documentation but found nothing about the
> runtime status of tomcat. how can i find out if tomcat is already
> running?
>
> thanks for any responses,
>
> thomas.
>

Re: Mapping with InvokerServlet in Tomcat 3.2.1

[Jeff Kilbride]

I was under the impression that to change the default servlet mapping, you
change the 'prefix' setting of org.apache.tomcat.request.InvokerInterceptor
in your server.xml file. I haven't needed to do it, though, so I don't know
if this is correct.

Give it a try.

Thanks,
--jeff

- Original Message -
From: "Andy Raffle" <[EMAIL PROTECTED]>
To: "Tomcat User" <[EMAIL PROTECTED]>
Sent: Saturday, June 23, 2001 12:02 AM
Subject: Mapping with InvokerServlet in Tomcat 3.2.1


> I've spent quite a few hours trying to migrate an existing Jserv 1.0
> servlet configuration to Tomcat 3.2.1, and I only have one question...
>
> Where is org.apache.tomcat.servlets.InvokerServlet ??
>
> According to the cvs, it was dropped a year ago, and yet in all the
> places that I've seen where it shows you how to change the servlet
> mapping (eg. from /servlet/* to /jbin/* in my case), it says to do
> this...
>
> 
> 
> invoker
> 
> 
> org.apache.tomcat.servlets.InvokerServlet
> 
> 
>
> 
> 
> invoker
> 
> 
> /jbin/*
> 
> 
>
> For me, this doesn't work. I can do something specific, such as:
>
> 
> 
> /jbin/xyz
> 
> 
> xyz
> 
> 
>
> That works. But the generic doesn't. I can't find InvokerServlet
> anywhere in the source or binaries, nor can I find any mention of the
> ...servlets package it's in. So it's either accidentally missing, or
> only the name is there for backwards compatibility, or the remapping
> mechanism is fundamentally broken. Or I've missed something!
>
> And yes, I have the directory structure right.
>
> I'm not looking for specific advice, unless it differs from the above.
> But it would be nice if someone who is 'in the know' would explain (a)
> where the InvokerServlet has gone, and (b) how the example above (almost
> verbatim from the FAQ) is ever supposed to work!
>
> Cheers!
>
> Andy.
>

Re: omegacms, thanks and a new problem: compile errors

[Jeff Kilbride]

Hi Vinny,

The "<%=" opening jsp tag is for single expression evaluation, while the
"<%" is for scriptlets. So, you could do this two ways:

<% out.print("!"); %>

or

<%= "!" %>

As you've seen, the "<%=" wraps whatever's inside it in an "out.print()"
statement -- so you're own "out.print()" is redundant in this case.

--jeff

- Original Message -
From: "Vinny" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, June 19, 2001 9:27 AM
Subject: Re: omegacms, thanks and a new problem: compile errors


> Thank you very much, I had to explicitly put the omega jar  and the
> directory containing the property file in my unix shell's classpath.
> Now the problem I'm running into is that some pages are getting compile
> error messages.
>
>
>
>
>   2001-06-19 12:07:40 - Ctx( /omega ): JasperException: R( /omega +
> /admin/hello.jsp + null) Unable to compile class for
>
JSP/usr/local/jakarta-tomcat-3.2.2/work/localhost_8080%2Fomega/_0002fadmin_0
002fhello_0002ejsphello_jsp_0.java:61:
> Incompatible type for method. Can't convert void to char[].
>  out.print( out.print("!") );
>
>
>
> the hello.jsp file contains:
>
>
> Hello World
> <%=out.print("!") %>
> 
>
>
>
> what's going on?
> Thanks again in advance.
>
>
>
>
>
>
> Randy Layman wrote:
>
> > If you're on UNIX, add it to TOMCAT_HOME/classes, if you're on NT
> > you'll need to modify the tomcat.bat file so that Tomcat adds this
directory
> > to its automatically built classpath.  Another option it to add the
> > conf.properties to the CLASSPATH environment variable.
> >
> > In either case, there are some ways that you can request resources
> > that cause the wrong class loader to be used, which it seems is
happening
> > here.  Since its closed source, there really is no way around this.
> >
> > Randy
> >
> >
> >>-Original Message-
> >>From: Vincent Stoessel [mailto:[EMAIL PROTECTED]]
> >>Sent: Tuesday, June 19, 2001 11:17 AM
> >>To: Tomcat Users
> >
> >>
>
>
>

Re: xtags: Unable to load class

[Jeff Kilbride]

Where did you put the taglib tld file and the taglib jar file? What does
your web.xml look like? The order of the definitions in web.xml is also
important. When I first set up my taglibs, I had the  definition in
my web.xml at the end of the file after a  definition
and it didn't work.

Check out tomcat.mslinn.com for an explanation of the web.xml file.

--jeff

- Original Message -
From: "William C. Robertson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, June 18, 2001 10:47 PM
Subject: xtags: Unable to load class


>
>
> -Original Message-
> From: William C. Robertson [mailto:[EMAIL PROTECTED]]
> Sent: Monday, June 18, 2001 4:37 PM
> To: [EMAIL PROTECTED]
> Subject: xtags: Unable to load class
>
>
> Has anyone else had a problem with taglibs?  I have everything configured
> correctly (pretty sure) and I get the following error when using any of
the
> tomcat taglibs (except for the example tagliv which can with tomcat and
> works fine):
>
> Error: 500
> Location: /examples/init_onlyImports.jsp
> Internal Servlet Error:
>
> org.apache.jasper.compiler.CompileException:
> /usr/local/jakarta-tomcat-3.2.2/webapps/examples/init_onlyImports.jsp(4,0)
> Unable to load class org.apache.taglibs.xtags.tags.ParseTag
> at
> org.apache.jasper.compiler.TagBeginGenerator.init(TagBeginGenerator.java,
> Compiled Code)
>
> I hope the problem is obvious to someone.  Thanks..
>

Re: Apache Default Document is .jsp?

[Jeff Kilbride]

You could also have an index.html that uses the meta-refresh tag to
automatically redirect to your login.jsp. This might be a lot easier than
mod_rewrite.

Either way, I would turn off directory browsing in Apache, unless that's
what you really want.

--jeff

- Original Message -
From: "Jason Koeninger" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, June 18, 2001 6:25 PM
Subject: Re: Apache Default Document is .jsp?


> Dig through the documentation on mod_rewrite and/or
> look at the Redirect command for Apache.  One or both
> of those two should be capable of accomplishing what
> you want.
>
> Best Regards,
>
> Jason Koeninger
> J&J Computer Consulting
> http://www.jjcc.com
>
> On Mon, 18 Jun 2001 17:40:02 -0700, Scott Jones wrote:
>
> >Hello,
> >
> >I'm getting ready to setup tomcat and Apache on seperate machines.
Before
> >getting started on that project, on my development machine, I set the
> >default "DocumentRoot" for apache to a different directory (for static
> >content) than my webapp (which will eventually sit on a different
machine).
> >
> >I'd like to have my "login.jsp" be my default document, but was only able
to
> >get it to work by putting a "dummy" login.jsp in the HTML directory...
> >Otherwise, Apache would just show a normal index of the directory...
> >
> >Is this the only way to get this to work?  Or am I missing somthing?
BTW,
> >I'm on Tomcat 3.2.1 and Apache 1.3.19...
> >
> >Thanks for any ideas.
> >
> >Cheers,
> >
> >Scott
> >
>
>
>

Re: request and jsp:include

[Jeff Kilbride]

You're missing an "=" in your jsp:include statement. Try this:

" />

You have it correct in your second example, which is why that one works.

Thanks,
--jeff

- Original Message - 
From: "Antoni Reus" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, June 18, 2001 4:58 AM
Subject: request and jsp:include


> Hi, first of all I would like to excuse for my poor english.
> 
> I'm using tomcat 3.2.2 in a Win NT 4
> 
> I'm trying to do some kinda printing template, so calling de template (
> imprimir.jsp )
> with a parameter for the page to show it whould display it in a
> printer friendly way.
> 
> In imprimir.jsp is something like this:
> 
> 
> ...
> ...
> " />
> ...
> ..
> 
> 
> when I call localhost:8080/imprimir.jsp?pagina=listado.jsp
> I get a tomcat exception when parsing imprimir.jsp, saying that
> pagina has no value.
> 
> But if I write imprimr.jsp  this way:
> 
> 
> ...
> ...
> <%! String pagina = request.getParameter("pagina");
> %>
> 
> ...
> ..
> 
> 
> it works!!
> 
> Someone could explain this???
> 

Re: Tomcat hangs if I refer to a context that doesn't exist

[Jeff Kilbride]

Actually, yes -- read the release notes for 3.2.1 that come with the
distribution.

--jeff

- Original Message -
From: "Jeff Trent" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Wednesday, June 13, 2001 5:30 AM
Subject: Re: Tomcat hangs if I refer to a context that doesn't exist


Actually, no - otherwise, I wouldn't have asked.  I did, however, find my
answer in the bugzilla database.  For those of you who are wondering, this
problem occurs under 3.2.1 if you remove the ROOT context / webapp.

-jeff
  - Original Message -
  From: Thomas Bezdicek
  To: [EMAIL PROTECTED]
  Sent: Wednesday, June 13, 2001 4:03 AM
  Subject: AW: Tomcat hangs if I refer to a context that doesn't exist


  Hi,

  you find it in the tomcat-documentation very easy.

  regards, tom
-Ursprüngliche Nachricht-
Von: Jeff Trent [mailto:[EMAIL PROTECTED]]
Gesendet: Mittwoch, 13. Juni 2001 06:40
An: [EMAIL PROTECTED]
Betreff: Tomcat hangs if I refer to a context that doesn't exist


If I refer to a nonexistant webapp, I find that Tomcat pins my CPU
(doesn't really hang).  I need to stop tomcat and restart for the CPU to
return to normal.  I couldn't find any information on this problem in the
archives.  Anybody else see this or know what the problem might be?  I'm
using 3.2.1 on NT2000.

thanks,
jeff

Re: mod_jk and mod_perl

[Jeff Kilbride]

I would go with option #1 -- build apache with mod_perl statically and
mod_so, then load mod_jk dynamically. I think you're going to have a lot of
work trying to get mod_jk to compile statically with apache, because the
developers haven't released an "easy" static version. You'd have to
integrate the mod_jk code into the apache source yourself.

The reason you can't use perl 5.005 may be your apache version (or apxs
version). You may also be able to switch to an older version of apache to
compile mod_perl as a DSO.

The other option is to upgrade perl. It's really not that hard and shouldn't
cause any major problems. You may want to try that first. (personally, I
think it's easier to upgrade perl than to install apache for the first
time...)

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, June 05, 2001 5:12 PM
Subject: RE: mod_jk and mod_perl


> We're running perl 5.005_03 on Solaris 2.6/sparc.  When I go to build
> mod_perl as a DSO, I get:
>
> * ERROR *
>
>   Your current configuration will most likely trigger core dumps,
> suggestions:
>*) Do not configure mod_perl as a DSO
>*) Upgrade your Perl version to 5.6.0 or higher (w/ -Ubincompat5005)
>*) Configure Perl with -Uusemymalloc (not recommended for performance)
>
>
> * ERROR *
>
> We're not in a position to upgrade perl or reconfigure perl.  So I can't
> build mod_perl as a DSO.
>
> My options are (in order of preference):
> - build apache with mod_perl (statically) and mod_so.  Then to load mod_jk
> with apxs.
> - build apache with mod_perl and mod_jk statically.
> - run two apache instances.  The first one will have mod_so and mod_jk and
> will handle requests for servlets and static pages.  The second will have
> mod_perl and will handle perl requests.
>
> If you, or anyone else, has any advice, I'd be stoked.
>
> Thanks,
> * * * John
>
>
> -Original Message-
> From: Jeff Kilbride [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, June 05, 2001 4:53 PM
> To: [EMAIL PROTECTED]
> Subject: Re: mod_jk and mod_perl
>
>
> I've run both mod_jk and mod_perl as dynamically linked modules in the
past,
> but not statically. The dynamic modules were very easy to get running.
>
> Any reason it has to be statically compiled?
>
> Thanks,
> --jeff
>
> - Original Message -
> From: <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, June 05, 2001 3:48 PM
> Subject: mod_jk and mod_perl
>
>
> > Hello,
> >
> > I'm trying to build a statically linked version of apache v1.3.19 with
> > mod_jk (from tomcat v3.2.1) and mod_perl v1.25 for Solaris 2.6/sparc.
Has
> > anyone successfully done this before?  I can build apache with mod_jk
and
> > apache with mod_perl, but I haven't been able to build apache with both
> > modules.  I've searched this mailing list and several others and have
made
> > no progress.  My next step is to deconstruct the mod_jk makefile to fit
> the
> > source into $APACHE_SRC/src/modules/jk.  Any hints are appreciated.
> >
> > Thanks,
> > * * * John Ubante
> >
>

Re: mod_jk and mod_perl

[Jeff Kilbride]

I've run both mod_jk and mod_perl as dynamically linked modules in the past,
but not statically. The dynamic modules were very easy to get running.

Any reason it has to be statically compiled?

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, June 05, 2001 3:48 PM
Subject: mod_jk and mod_perl


> Hello,
>
> I'm trying to build a statically linked version of apache v1.3.19 with
> mod_jk (from tomcat v3.2.1) and mod_perl v1.25 for Solaris 2.6/sparc.  Has
> anyone successfully done this before?  I can build apache with mod_jk and
> apache with mod_perl, but I haven't been able to build apache with both
> modules.  I've searched this mailing list and several others and have made
> no progress.  My next step is to deconstruct the mod_jk makefile to fit
the
> source into $APACHE_SRC/src/modules/jk.  Any hints are appreciated.
>
> Thanks,
> * * * John Ubante
>

Re: Tomcat dying: solution!!

[Jeff Kilbride]

I'm glad to hear this worked!

Thanks,
--jeff

- Original Message -
From: "Joe Howes" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, June 04, 2001 10:18 PM
Subject: Tomcat dying: solution!!


> Don't know if this will help any of you who have had tomcat die
> mysteriously, but it sure seems to have stabilized my installs.
>
> I followed Jeff's advice and upped the max_threads (server.xml file, see
> http://jakarta.apache.org/tomcat/tomcat-3.2-doc/uguide/tomcat_ug.html
> and grep for max_threads.)
>
> My jakarta instances on three machines would die on a very regular
> basis...each machine at least one or two times an hour.  I specified all
> the options for the PoolTcpConnector (actually just used exactly the
> settings from the above web page) and none of them have died since.
> Hopefully when I get in tomorrow they'll still be up and running :)  But
> this seems to have done it.
>
>
> - Joe
>
>
>
> > Jeff Kilbride wrote:
> > >
> > > I seem to remember something from the tomcat-dev list about 3.2.x
dying
> less
> > > than gracefully if the max_threads parameter for PoolTCPConnector is
> ever
> > > exceeded. I believe the default is 50, so if you're ever hitting more
> than
> > > 50 concurrent threads, maybe this is the problem.
> > >
> > > If you're using Tomcat with Apache, try upping your max_threads
> parameter to
> > > match the max number of child processes your Apache installation
allows
> > > (MaxClients in httpd.conf). If you're running standalone, up
max_threads
> to
> > > a reasonable number you don't think you'll hit. If I remember
correctly,
> > > this solved the problem for someone else -- or at least significantly
> > > prolonged Tomcat's life cycle.
> > >
> > > I'd really be interested in hearing if this helps.
> > >
> > > Thanks,
> > > --jeff
> > >
> > > - Original Message -
> > > From: "Hunter Hillegas" <[EMAIL PROTECTED]>
> > > To: "Tomcat User List" <[EMAIL PROTECTED]>
> > > Sent: Friday, June 01, 2001 12:20 PM
> > > Subject: 3.2.2 Dies After Prolonged Use...
> > >
> > > > You may remember my posts about Tomcat dying on me... Well I
upgraded
> to
> > > > 3.2.2 and it is still happening.
> > > >
> > > > It only seems to happen after prolonged periods (lots of hits)...
> > > >
> > > > I increased the heap to 256MB with a max of 512MB. We're not using
> > > sessions
> > > > on the site and the session timeout is set to 5 minutes anyway...
> > > >
> > > > What could be going on?
> > > >
> > > > Hunter
> > > >
> >
>

Re: 3.2.2 Dies After Prolonged Use...

[Jeff Kilbride]

Check out the User's Guide:

http://jakarta.apache.org/tomcat/tomcat-3.2-doc/uguide/tomcat_ug.html

and do a "find" in your browser for "max_threads".

Thanks,
--jeff

- Original Message -
From: "Joe Howes" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, June 02, 2001 10:03 AM
Subject: Re: 3.2.2 Dies After Prolonged Use...


> Thanks for the suggestion, Jeff.
>
> Stupid question...I've grepped all over hell and back and I can't find a
> max_threads param in any config files :)  Where do I find it?
>
> - Joe
>
> Jeff Kilbride wrote:
> >
> > I seem to remember something from the tomcat-dev list about 3.2.x dying
less
> > than gracefully if the max_threads parameter for PoolTCPConnector is
ever
> > exceeded. I believe the default is 50, so if you're ever hitting more
than
> > 50 concurrent threads, maybe this is the problem.
> >
> > If you're using Tomcat with Apache, try upping your max_threads
parameter to
> > match the max number of child processes your Apache installation allows
> > (MaxClients in httpd.conf). If you're running standalone, up max_threads
to
> > a reasonable number you don't think you'll hit. If I remember correctly,
> > this solved the problem for someone else -- or at least significantly
> > prolonged Tomcat's life cycle.
> >
> > I'd really be interested in hearing if this helps.
> >
> > Thanks,
> > --jeff
> >
> > - Original Message -
> > From: "Hunter Hillegas" <[EMAIL PROTECTED]>
> > To: "Tomcat User List" <[EMAIL PROTECTED]>
> > Sent: Friday, June 01, 2001 12:20 PM
> > Subject: 3.2.2 Dies After Prolonged Use...
> >
> > > You may remember my posts about Tomcat dying on me... Well I upgraded
to
> > > 3.2.2 and it is still happening.
> > >
> > > It only seems to happen after prolonged periods (lots of hits)...
> > >
> > > I increased the heap to 256MB with a max of 512MB. We're not using
> > sessions
> > > on the site and the session timeout is set to 5 minutes anyway...
> > >
> > > What could be going on?
> > >
> > > Hunter
> > >
>

Re: 3.2.2 Dies After Prolonged Use...

[Jeff Kilbride]

I seem to remember something from the tomcat-dev list about 3.2.x dying less
than gracefully if the max_threads parameter for PoolTCPConnector is ever
exceeded. I believe the default is 50, so if you're ever hitting more than
50 concurrent threads, maybe this is the problem.

If you're using Tomcat with Apache, try upping your max_threads parameter to
match the max number of child processes your Apache installation allows
(MaxClients in httpd.conf). If you're running standalone, up max_threads to
a reasonable number you don't think you'll hit. If I remember correctly,
this solved the problem for someone else -- or at least significantly
prolonged Tomcat's life cycle.

I'd really be interested in hearing if this helps.

Thanks,
--jeff

- Original Message -
From: "Hunter Hillegas" <[EMAIL PROTECTED]>
To: "Tomcat User List" <[EMAIL PROTECTED]>
Sent: Friday, June 01, 2001 12:20 PM
Subject: 3.2.2 Dies After Prolonged Use...


> You may remember my posts about Tomcat dying on me... Well I upgraded to
> 3.2.2 and it is still happening.
>
> It only seems to happen after prolonged periods (lots of hits)...
>
> I increased the heap to 256MB with a max of 512MB. We're not using
sessions
> on the site and the session timeout is set to 5 minutes anyway...
>
> What could be going on?
>
> Hunter
>

Re: ** JVM and Processes

[Jeff Kilbride]

Hi Adam,

No, the garbage collector runs as a low priority background process and, on
a lightly loaded server, may never get called because the server's not using
enough resources to warrant it. I really wouldn't worry about it too much
and I would definitely avoid killing threads individually, especially since
you're now utilizing a Pool connector. (you don't want to kill threads that
are marked as available in the pool...) The min_spare_threads and
max_spare_threads settings are supposed to take care of cleaning up any
"extra" unused threads that are laying around.

I think the best benefit you could do yourself would be to upgrade your 3.1
version of Tomcat to the newly released 3.2.2 final to take advantage of
upgrades and bug fixes.

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, June 01, 2001 11:03 AM
Subject: Re: ** JVM and Processes


> Jeff,
>
> Thanks a bunch. Your answer appears to be the best so far. I have
> implemented the PoolTCPConnector in the server xml file and it appears to
> be limiting the number of threads as it should. However, something that
has
> been happening (even before switching to PoolTCPConnector) is that when
> running multiple java servlets the threads stay alive long after they
> should have died or been garbage collected. Even after a long wait, the
> only way (apparently) to get rid of them is to go through and kill them
one
> at a time. Is there a setting somewhere that is telling the java threads
to
> stay alive indefinitely?
>
> Thanks for your help,
>  - Adam
>
>
> At 10:34 AM 6/1/2001 -0700, you wrote:
> >When Java first came to the Linux platform (via the Blackdown port),
> >green-threads were the only option. Native threads took a little longer
to
> >implement, but are a much better option for the reasons listed in the
> >previous message. So, I would recommend avoiding green-threads unless you
> >have a specific reason for using them.
> >
> >A lot of people freak out when they see the number of "processes" being
> >reported by ps or top, without realizing that these are merely threads
and
> >not full-blown processes. If you have a lightly loaded Tomcat, you can
tune
> >down the number of threads being spawned by using the max_threads,
> >max_spare_threads, and min_spare_threads parameters of the
PoolTCPConnector
> >in your server.xml file. For an example of this, take a look at the
tomcat
> >user's guide:
> >
> >http://jakarta.apache.org/tomcat/tomcat-3.2-doc/index.html
> >
> >Do a "find" in your web browser for "max_threads". I use this to limit
the
> >number of ajp12 threads and maximize ajp13 threads -- because I'm using
> >ajp13 for my servlets and ajp12 only for startup/shutdown of Tomcat.
> >
> >Conversely, if you have a heavily loaded Tomcat, you should use these
> >parameters to increase performance.
> >
> >Thanks,
> >--jeff
> >
> >- Original Message -
> >From: "Michael Jennings" <[EMAIL PROTECTED]>
> >To: <[EMAIL PROTECTED]>
> >Sent: Friday, June 01, 2001 9:32 AM
> >Subject: Re: ** JVM and Processes
> >
> >
> >RE: ** JVM and ProcessesMy understanding of green vs. native threads is
as
> >follows:
> >With native threads, an actual system thread is created when a Java
thread
> >is created.
> >On linux a system thread takes the form of another process, but one that
> >shares memory
> >etc. with another process. This is why if you create a program that
> >allocates 100 megs of memory,
> >then spins off 10 threads, it looks like 10 processes each taking up 100
> >megs of memory, when in
> >fact the amount of memory is 100 megs + 10*overhead for each thread (not
> >much more than 100 megs).
> >
> >On WIN32 systems, threads do not show up as separate processes, they are
> >separate threads of execution
> >inside the same process (essentially the same as the Linux implementation
> >with differences too subtle to care about)
> >
> >Green threads on the other hand use timers, signals, setjmp etc. voodoo
to
> >"simulate" threads within one process.
> >Essentially taking over the scheduling from the kernel.
> >
> >I believe the command-line option for green threads is simply "-green" as
in
> >java -green MyThreaddedApp
> >
> >If you have a multi-cpu system, green threads will only take advantage of
> >one cpu, whereas native threads
> >will use all the cpus on your system (that's the theory anyway)
> >
> >I've heard of problems with blocking I/O with green threads, but have no
> >first hand knowledge.
> >
> >Hope this helps.
> >-Mike Jennings
> >
> >  - Original Message -
> >  From: BARRAUD Valérie
> >  To: '[EMAIL PROTECTED]'
> >  Sent: Friday, June 01, 2001 9:01 AM
> >  Subject: RE: ** JVM and Processes
> >
> >
> >
> >
> >  http://java.sun.com/products/jdk/1.1/packs/native-threads/README
> >
> >-Message d'origine-
> >De: [EMAIL PROTECTED] [SMTP:[EMAIL PROTECTED]]
> >Date:   vendredi 1 juin 2001 17:46
> >À:  [EMAIL PROTECTED]
> >Objet:  RE: ** JVM and

Re: ** JVM and Processes

[Jeff Kilbride]

When Java first came to the Linux platform (via the Blackdown port),
green-threads were the only option. Native threads took a little longer to
implement, but are a much better option for the reasons listed in the
previous message. So, I would recommend avoiding green-threads unless you
have a specific reason for using them.

A lot of people freak out when they see the number of "processes" being
reported by ps or top, without realizing that these are merely threads and
not full-blown processes. If you have a lightly loaded Tomcat, you can tune
down the number of threads being spawned by using the max_threads,
max_spare_threads, and min_spare_threads parameters of the PoolTCPConnector
in your server.xml file. For an example of this, take a look at the tomcat
user's guide:

http://jakarta.apache.org/tomcat/tomcat-3.2-doc/index.html

Do a "find" in your web browser for "max_threads". I use this to limit the
number of ajp12 threads and maximize ajp13 threads -- because I'm using
ajp13 for my servlets and ajp12 only for startup/shutdown of Tomcat.

Conversely, if you have a heavily loaded Tomcat, you should use these
parameters to increase performance.

Thanks,
--jeff

- Original Message -
From: "Michael Jennings" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, June 01, 2001 9:32 AM
Subject: Re: ** JVM and Processes


RE: ** JVM and ProcessesMy understanding of green vs. native threads is as
follows:
With native threads, an actual system thread is created when a Java thread
is created.
On linux a system thread takes the form of another process, but one that
shares memory
etc. with another process. This is why if you create a program that
allocates 100 megs of memory,
then spins off 10 threads, it looks like 10 processes each taking up 100
megs of memory, when in
fact the amount of memory is 100 megs + 10*overhead for each thread (not
much more than 100 megs).

On WIN32 systems, threads do not show up as separate processes, they are
separate threads of execution
inside the same process (essentially the same as the Linux implementation
with differences too subtle to care about)

Green threads on the other hand use timers, signals, setjmp etc. voodoo to
"simulate" threads within one process.
Essentially taking over the scheduling from the kernel.

I believe the command-line option for green threads is simply "-green" as in
java -green MyThreaddedApp

If you have a multi-cpu system, green threads will only take advantage of
one cpu, whereas native threads
will use all the cpus on your system (that's the theory anyway)

I've heard of problems with blocking I/O with green threads, but have no
first hand knowledge.

Hope this helps.
-Mike Jennings

  - Original Message -
  From: BARRAUD Valérie
  To: '[EMAIL PROTECTED]'
  Sent: Friday, June 01, 2001 9:01 AM
  Subject: RE: ** JVM and Processes




  http://java.sun.com/products/jdk/1.1/packs/native-threads/README

-Message d'origine-
De: [EMAIL PROTECTED] [SMTP:[EMAIL PROTECTED]]
Date:   vendredi 1 juin 2001 17:46
À:  [EMAIL PROTECTED]
Objet:  RE: ** JVM and Processes

Randy,

Thanks for the advice. Could you be a little more specific, though,
about
how to use green threads instead of native threads and possibly
differences
between the two? Thanks.

 - Adam



At 10:59 AM 6/1/2001 -0400, you wrote:
>
>   Don't use ps - these are actually threads.  ps is showing them
as
>processes because that is what it does.  If you use green thread (as
opposed
>to the native threads you are using now), the display will go away, but
you
>will experience a slowdown (how much depends on your operating system
and
>other activity on the system).
>
>   Randy
>
>
>> -Original Message-
>> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
>> Sent: Friday, June 01, 2001 10:37 AM
>> To: [EMAIL PROTECTED]
>> Subject: ** JVM and Processes
>>
>>
>> Hi,
>>
>> For a particular web server we are running with Tomcat 3.1,
>> we are having
>> an issue with the java servlets that are running. What appears to be
>> happening is that each time a servlet is called from the web
>> site, a new
>> process is created to run the java program. When I view
>> processes with "ps
>> ax", I see dozens of instances of:
>> /usr/java/jdk1.3/bin/i386/native_threads/java
>>
>> It was briefly stated in Java Servlet Programming by Hunter &
>> Crawford, (c)
>> Oreilly that 'most servlet containers execute all servlets in
>> a single JVM
>> ... the exception being high-end containers that support
>> execution across
>> multiple backend servers...'
>>
>> We are only using 1 web server with an average weekly load of
>> a couple of
>> hundred visitors.
>>
>> Any ideas as to why we would be seeing so many identical
>> processes and if
>> so, how to modif

Re: Virtual Host Context Aliasing

[Jeff Kilbride]

What version of Tomcat is supposed to have this  tag?

Thanks,
--jeff

- Original Message -
From: "Christian Parpart" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, May 31, 2001 5:56 PM
Subject: Re: Virtual Host Context Aliasing


> On Friday 01 June 2001 02:40, you wrote:
> > I have been told from a collegue of mine that an alias tag can be used.
> > I have not tested this though, let me know if this works :)
> >
> >
> >  
> >...
> >
> >...
> >  
> >
>
> I've tested it without any success. Unfortunately.
> But exactly that's it what has been said in the tomcat-server-howto.
> Is it a bug? It must be
>
> Thanks,
> Christian Parpart
> http://www.surakware.net
>
> >
> > --Marcus
> >
> > On Thu, 31 May 2001, Jeff Kilbride wrote:
> > > Date: Thu, 31 May 2001 11:12:38 -0700
> > > From: Jeff Kilbride <[EMAIL PROTECTED]>
> > > Reply-To: [EMAIL PROTECTED]
> > > To: [EMAIL PROTECTED]
> > > Subject: Re: Virtual Host Context Aliasing
> > >
> > > Hi Daniel,
> > >
> > > I have the same problem -- wanting to alias more than one host name to
a
> > > single context. Unfortunately, there is no way to do this. I've heard
> > > that the 4.0 version of Tomcat may have some abilities for doing this,
> > > but it's not nearly as straightforward as Apache's ServerAlias
directive.
> > >
> > > For now, I've turned off zzz.net (using your example) in my DNS and am
> > > only serving www.zzz.net. I do a lot of connection pooling and other
> > > shared resources, so I can't afford to have 2 versions of all my
contexts
> > > running at the same time. This is a totally unsatisfactory solution,
but
> > > it's my only choice for the moment.
> > >
> > > Thanks,
> > > --jeff
> > >
> > > - Original Message -
> > > From: "Daniel Zen" <[EMAIL PROTECTED]>
> > > To: <[EMAIL PROTECTED]>
> > > Sent: Tuesday, May 29, 2001 8:55 AM
> > > Subject: Virtual Host Context Aliasing
> > >
> > > > I would think this would be a common question, but I couldn't find
it
> > > > documented, nor asked on this list.
> > > >
> > > > Very often domains are served from 2 urls (www.zzz.net & zzz.net)
with
> > > > the same functionality. When I configure my virtual hosts in
Apache's
> > >
> > > httpd.conf
> > >
> > > > this is easy:
> > > >
> > > > 
> > > >  ServerName www.zzz.net
> > > >  ServerAlias zzz.net
> > > >  DocumentRoot /home/httpd/html/zzz
> > > >  
> > > >   Options None
> > > >   Deny from all
> > > >  
> > > >  JkMount /*.jsp ajp13
> > > >  JkMount /servlet/* ajp13
> > > > 
> > > >
> > > > The following properly placed in server.xml creates 2 seperate
contexts
> > >
> > > for
> > >
> > > > the same set of servlets and JSPs. Functional, but a little
wasteful.
> > > >
> > > >   
> > > > > > > crossContext="true" debug="0" reloadable="true"
trusted="false"
> > > > /> 
> > > >
> > > >   
> > > > > > > crossContext="true" debug="0" reloadable="true"
trusted="false"
> > > > /> 
> > > >
> > > > Now, I how do I do an alias Context in Tomcat's server.xml so that
> > > > there
> > >
> > > is
> > >
> > > > only one Host/Context with multiple names??
> > > >
> > > > Thank you in advance.
> > > >
> > > > Daniel Zen
>

Re: How to debug a missing servlet error?

[Jeff Kilbride]

Hi Chris,

Tomcat should recognize "/servlet/briefXSL" without the explicit
servlet-mapping you are using -- but I don't know if that is what's causing
your problem. For all my servlets, I have the following type of entry:


briefXSL
com.smartbrief.BriefXSL.Servlet


The default Invoker automatically sets up "/servlet/" as a mapping for all
your defined servlets. So, the above should be enough to get
"/servlet/briefXSL" to pull up correctly -- without the servlet-mapping you
have below. Maybe the explicit servlet-mapping you are doing is somehow
messing with the default Invoker on Linux, but that's only speculation...

Thanks,
--jeff

- Original Message -
From: "Chris McNeilly" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, May 31, 2001 1:15 PM
Subject: How to debug a missing servlet error?


> Hi,
>
> I have a development environment that works correctly (Win 98), but when
> I move the code over to my QA environment (Linux) tomcat can no longer
> find the servlet.  I have a web.xml file in the Web-Inf directory that
> has the following:
>
> 
> 
> 
> briefXSL
> 
> 
> com.smartbrief.BriefXSLServlet
> 
> 
> 
> briefXSL
> /servlet/briefXSL
> 
>
> 
>
> Tomcat receives the request from apache, but doesn't know what to do
> with it and spits back a 404.  It's almost as if tomcat isn't reading
> the web.xml file at all.
>
> Thanks,
>
> Chris
>

Re: Tomcat 3.2.2 bug ???

[Jeff Kilbride]

I think I saw something about this on the tomcat-dev mailing list.

Try changing your 404.html file to 404.jsp. You don't need to change the
file at all, just rename it with the new extension. Then change the
 tag to /404.jsp.

Thanks,
--jeff

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, May 31, 2001 1:01 PM
Subject: Tomcat 3.2.2 bug ???


> When i request a non-existing jsp page, the server crash...
> web.xml
> 
>404
>/404.html
>  
>
> I think the problem is  web.xml file in  tag.. It do not
accept
> static error pages... Does anybody know the solution?
>

Re: Apache + Tomcat with mod_jk and Virtual Hosts

[Jeff Kilbride]

Yes. See the mail archive for a generic version of my setup:

http://mikal.org/interests/java/tomcat/archive/view?mesg=24718
http://mikal.org/interests/java/tomcat/archive/view?mesg=24420
http://mikal.org/interests/java/tomcat/archive/view?mesg=24256

Thanks,
--jeff

- Original Message - 
From: "Adrian Almenar" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, May 31, 2001 7:03 AM
Subject: Apache + Tomcat with mod_jk and Virtual Hosts


> Please i need help with this can anyone anwser me
> about this issue ??
> 
> 
> With Apache and tomcat working with mod_jk:
> 
> Its possible to map every virtual host (On Apache)
> to a different webapp on tomcat ???
> 
> I.E.
> 
> Apache Virtual Host: 123.myhost.com ip: 10.0.0.2
> Tomcat Webbapp : tomcat33\webapps\123
> 
> Apache Virtual Host: 789.myhost.com ip: 10.0.0.2
> Tomcat Webbapp : tomcat33\webapps\789
> 
> thanks in advance !!!
> 

Re: Virtual Host Context Aliasing

[Jeff Kilbride]

Hi Daniel,

I have the same problem -- wanting to alias more than one host name to a
single context. Unfortunately, there is no way to do this. I've heard that
the 4.0 version of Tomcat may have some abilities for doing this, but it's
not nearly as straightforward as Apache's ServerAlias directive.

For now, I've turned off zzz.net (using your example) in my DNS and am only
serving www.zzz.net. I do a lot of connection pooling and other shared
resources, so I can't afford to have 2 versions of all my contexts running
at the same time. This is a totally unsatisfactory solution, but it's my
only choice for the moment.

Thanks,
--jeff

- Original Message -
From: "Daniel Zen" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, May 29, 2001 8:55 AM
Subject: Virtual Host Context Aliasing


> I would think this would be a common question, but I couldn't find it
> documented, nor asked on this list.
>
> Very often domains are served from 2 urls (www.zzz.net & zzz.net) with the
> same functionality. When I configure my virtual hosts in Apache's
httpd.conf
> this is easy:
>
> 
>  ServerName www.zzz.net
>  ServerAlias zzz.net
>  DocumentRoot /home/httpd/html/zzz
>  
>   Options None
>   Deny from all
>  
>  JkMount /*.jsp ajp13
>  JkMount /servlet/* ajp13
> 
>
> The following properly placed in server.xml creates 2 seperate contexts
for
> the same set of servlets and JSPs. Functional, but a little wasteful.
>
>   
> crossContext="true" debug="0" reloadable="true" trusted="false" />
>   
>
>   
> crossContext="true" debug="0" reloadable="true" trusted="false" />
>   
>
> Now, I how do I do an alias Context in Tomcat's server.xml so that there
is
> only one Host/Context with multiple names??
>
> Thank you in advance.
>
> Daniel Zen
>

Re: Setting up Virtual Hosts

[Jeff Kilbride]

Yes:

NameVirtualHost 111.222.333.444



server.xml:


This is how I have it set up and it works for about 10 domains.

--jeff


> From: Glen Eustace <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Wed, 23 May 2001 05:28:13 GMT
> To: [EMAIL PROTECTED]
> Subject: Re: Setting up Virtual Hosts
> 
>> Apache:
>> 
> 
>> server.xml:
>> 
> 
>> In your config, you're using an IP in the Apache VirtualHost directive
> and a
>> hostname in your server.xml config. I have a feeling that may be causing
>> your problems. Try changing your Apache config to use the host name in
> the
>> VirtualHost directive, rather than the IPAddress and see if that makes a
>> difference.
> 
> Jeff, can you please confirm that you have the following directive ( with
> a different IP number of course )
> 
> NameVirtualHost 210.55.214.169
> 
> in your apache config, and that you have multiple hosts using that same
> IP number, and that these hosts are the same ones in you tomcat
> server.xml config. My experimentation would still suggest that tomcat is
> not doing named virtual hosts.
> 
> Glen.
> 

Re: 3.2.1 Dies

[Jeff Kilbride]

It's not necessary to use nohup, but it's a good idea to redirect your
standard out and standard error to a logfile when starting up. I use the
following line in my startup script called "jkup" which I placed in
/usr/local/bin:

$TOMCAT_HOME/bin/startup.sh &>/var/log/tomcat/startup.log

My logs directory in $TOMCAT_HOME is a soft link to /var/log/tomcat, where I
keep all my Tomcat related logs.

--jeff

> From: Boris Niyazov <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Mon, 21 May 2001 15:59:43 -0400 (EDT)
> To: [EMAIL PROTECTED]
> Subject: Re: 3.2.1 Dies
> 
> Why use nohup if you can configure tomcat to log into a file?
> 
>  path="logs/tomcat.log"
> verbosityLevel = "DEBUG"
> />
> 
> *
> * Boris NiyazovPh:  212-854-4094  Fax: 212-854-1749 *
> * Systems Manager  Email: [EMAIL PROTECTED] *
> * Columbia Law School  URL: http://www.law.columbia.edu *
> *
> 
> 
> 
> 
>> 
>> STDOUT & STDERR messages go to nohup.log. In your case you might be loosing
>> those messages as you logged off (they will be sent to /dev/null - trash).
>> 
>> Sri
>> 
>> At 10:54 AM 05/21/2001 -0700, Hunter Hillegas wrote:
>>> What does running with nohup do for you?
>>> 
>>> I usually start Tomcat using tomcat.sh start and then just log out...
>>> 
>>> Hunter
>>> 
 From: Srinadh Karumuri <[EMAIL PROTECTED]>
 Date: Mon, 21 May 2001 13:43:53 -0400
 To: Hunter Hillegas <[EMAIL PROTECTED]>, Tomcat User List
 <[EMAIL PROTECTED]>
 Subject: Re: 3.2.1 Dies
 
 - I am running the tomcat using 'nohup' (UNIX command) on Solaris. My
 OutOfMemory errors were logged in nohup.log
>>> 
>> 
>> Srinadh Karumuri
>> Senior Programmer/Analyst
>> Business Apps.
>> BBN Technologies (Verizon)
>> Ph:(617)873-2841
>> 
> 

Re: Setting up Virtual Hosts

[Jeff Kilbride]

Hi Glen,

It definitely works with name-based hosts, because that's how I do it.
However, I set mine up like this:

Apache:


server.xml:


In your config, you're using an IP in the Apache VirtualHost directive and a
hostname in your server.xml config. I have a feeling that may be causing
your problems. Try changing your Apache config to use the host name in the
VirtualHost directive, rather than the IPAddress and see if that makes a
difference.

Thanks,
--jeff

> From: Glen Eustace <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Mon, 21 May 2001 08:26:22 GMT
> To: [EMAIL PROTECTED]
> Subject: Re: Setting up Virtual Hosts
> 
> 
>>> http://mikal.org/interests/java/tomcat/index.jsp
> 
>>> Look for Virtual Host or my name.
> 
>> Will do.  Thanks. This is very frustrating.
> 
> Having browsed around in the archives, I have come to the ( possibly
> erroeous ) conclusion that tomcat does not support NAMED virtual hosts,
> but that they are actually IP based.
> 
> Can someone confim whether tomcat really does use the Host: header to
> determine the virtual host or is it resolving the host in the URL to an
> IP number and then finding a host definition that matches it.
> 
> Thanks.
> -- 
> Glen Eustace,
> GodZone Internet Services, a div. of AGRE Enterprises Ltd.,
> P.O. Box 8020, Palmerston North, New Zealand 5301
> Tel/Fax: +64 6 357 8168, Mob: +64 21 424 015
> E-mail: [EMAIL PROTECTED] Website: http://www.godzone.net.nz
> 

Re: Setting up Virtual Hosts

[Jeff Kilbride]

I'm not sure how Apache forwards the host info to Tomcat. If you're using an
IP Address in your VirtualHost directive in Apache, then Apache may be
sending the IPAddress to Tomcat as opposed to the hostname specified by the
ServerName attribute. So, if you're trying to catch host names in your
server.xml file, rather than IPAddresses, it may not work. However, this is
all speculation and I don't know if that's how it really works. It would
make sense, though.

The docBase is where Tomcat serves all your files from for that particular
webapp. Check out the "Developing applications with Tomcat" Howto in the
docs for an example of setting up source and deployment directories.

--jeff

> From: Glen Eustace <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> Date: Mon, 21 May 2001 05:16:01 GMT
> To: [EMAIL PROTECTED]
> Subject: Re: Setting up Virtual Hosts
> 
>> I don't know if it makes a difference or not, but I've always had the
> same
>> host names in my Apache VHost config and my server.xml file. In your
> config,
>> you're using an IP Address in Apache and a name in server.xml. You might
> try
>> using the IP in server.xml instead.
> 
> All my vhosts use the same IP number as they are all named hosts.  Using
> the IP number in the VirtualHost header just saves a DNS lookup and make
> starting the server faster.  The host: header is matched against the
> ServerName attribute.
> 
>> Nothing else really jumps out at me. For an example of my config, search
> the
>> mail archive at:
> 
>> http://mikal.org/interests/java/tomcat/index.jsp
> 
>> Look for Virtual Host or my name.
> 
> Will do.  Thanks. This is very frustrating.
> 
> Another thought I had was whether the docBase directory must have some
> special format.  The webapps one in the tomcat directory has a bunch of
> .war files ( which I have no idea what they are for ).
> 
> Glen.
>