As Craig already said, it's not good to generalize.
The only way to find out, is to test. (And always be prepared
that the result may change in the next release of the vm)
If the garbage collector works as you describe, it's quite easy
to improve it in a way that it does the opposite of your
There happens a NullPointerException in line 47 of
eshop.share.LoginCommand.
Without the source of that class (or at least a relevant snippet)
we can't help you much.
-Original Message-
From: Lindomar [mailto:[EMAIL PROTECTED]]
Sent: Monday, February 03, 2003 6:03 PM
To: Tomcat Users
We use the xvfb to achieve that goal. (It's
typically part of the linux distribution)
With that it works quite transparently without
any additional toolkit.
The only thing we have to change in the tomcat
installation is to set an env var DISPLAY that
points to te virtual frame buffer.
In gregorian calendar aplies following rule
isLeapYear = (year mod 4 = 0) and ((not year mod 100 = 0) or (year mod 400 = 0))
2000 mod 4 = 0 - true
not 2000 mod 100 = 0 - false
2000 mod 400 = 0 - true
true and (false or true) - true
-Original Message-
From: Felipe Schnack
There have been several suggestions to achieve this:
- use apache
- NAT with ipchain/iptables
(A german article for iptables: http://www.3plus4software.de/news/20020617.html)
- bindd
- authbind
http://packages.debian.org/unstable/utils/authbind.html
a discussion about authbind vs. bindd
I can see just one attachment with the unsubscribe
info of the tomcat list but no web.xml. (Unless my
mailreader is doing something strange)
-Original Message-
From: Heger, Karlheinz [mailto:[EMAIL PROTECTED]]
Sent: Thursday, February 13, 2003 12:54 PM
To: '[EMAIL PROTECTED]'
Because the API doc says so :) (Sorry, couldn't resist)
I'm not aware of the rationale behind that.
(If there is any at all)
-Original Message-
From: Shapira, Yoav [mailto:[EMAIL PROTECTED]]
Sent: Thursday, February 13, 2003 4:18 PM
To: [EMAIL PROTECTED]
Subject:
Just a very wild guess:
Have you compared the date of the jsp file
with the corresponding java and class file?
Maybe the old class file is newer than the jsp file?
-Original Message-
From: Sven Köhler [mailto:[EMAIL PROTECTED]]
Sent: Friday, February 14, 2003 10:04 AM
To: [EMAIL
I guess you are using jdk 1.4 ?
AFAIK with that it's not allowed to import packageless classes.
You have to put Katalog and Menu in a package and import that.
-Original Message-
From: Terje Hopsø [mailto:[EMAIL PROTECTED]]
Sent: Friday, February 14, 2003 1:42 PM
To: [EMAIL
That doesn't sound like an error that is related to mod_jk
at all. It's either a missconfiguration of the basic apache
or there is already a server listening on port 80 (Another
Apache or IIS; you can verify this with netstat).
-Original Message-
From: Carlos Cajina [mailto:[EMAIL
You have an error in web.xml.
- Your tags don't follow the specified order
- you have a tag that is not defined in the dtd.
- You omitted a required tag
-Original Message-
From: Sanjeevkumar Cherengotil
[mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 18, 2003 1:15 PM
To: Tomcat
Without the source for TestConnectionServlet.java
it's hard to help.
-Original Message-
From: sophie seillier [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 18, 2003 6:37 PM
To: [EMAIL PROTECTED]
Subject: Internal Server Error nullPointerException
I don't know what you exacly mean with 'lively project'
but the last time I had to deal with the postgres driver
the development was very active. (The result of the
activity was not always optimal) Each time I found
a bug, it was resolved by the time I got enough
information to file a bug
It depends on your requirements.
For read only load balancing to achieve more performance
you might also consider to use replication:
http://gborg.postgresql.org/project/pgreplication/projdisplay.php
This way you could for exapmle setup 4 instances of
tomcat and 2 instances of postgreSQL where
The mosts posts in this list indicate that the server
option is not a good option. It's faster, but also
less stable.
Havn't seen anything about the latests JDK's in this
respect. So you may have to try and stress test it.
-Original Message-
From: Sven Köhler [mailto:[EMAIL
No, the ideal would be faster and more stable :)
Just some statements from Sun:
The Java HoSpot Client VM ... it is tuned for best
performance ... by reducing application start-up
time and memory footprint.
The Java HotSpot Server VM is designed for maximum
program execution speed.
You have to make shure that your script retstart_tomcat
sets and exports all needed environment variables before
calling ./startup.sh:
JAVA_HOME=/usr/local/java/jdk1.3.1
CATALINA_HOME=path to tomcat installation
CATALINA_BASE=path to tomcat instance or $CATALINA_HOME
# JAVA_OPTS='-client -v'
As you can see my personal preference is to reply first
then quote the relevant part of the original message.
(I prefer this, as posts in this format enable me
to see if the answer has enough info to care about the
question, in most cases I'm aware of the question anyway
as I follow the list
Just a side note:
If the sole reason for this jsp is the automatic check
then your example can be stripped down to:
SUCCESS
The rest is just to be a little bit more friendly
to a browser.
-Original Message-
From: Turner, John [mailto:[EMAIL PROTECTED]
Sent: Tuesday, February 25,
I would consider two monitors:
- One local that restarts tomcat if the process is not
alife anymore.
For this monitor ps can be enough. More sophisticated
checks should only be done if you are shure that you
want to automatically restart tomcat if this checks
fail.
- One remote that
Because the underlying classes sometimes cache a negative
response, so you have to restart tomcat to enable a new
lookup. (That's not specific to tomcat)
-Original Message-
From: Hannes Schmidt [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 26, 2003 3:02 AM
To: Tomcat Users List
It's the java implementation that does the caching, as java implements
the lookup on it's own and doesn't use the operating system functions
for that. (That doesn't mean that the operating system or the resolver
lib is not caching, but that is independent
of the java problem.)
The lookup
I second that.
Only the setting for networkaddress.cache.negative.ttl
makes sense as there is no ttl if you don't get a valid
response.
-Original Message-
From: Turner, John [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 26, 2003 2:17 PM
To: 'Tomcat Users List'
Subject: RE:
AFAIK tomcat has no own timeout of this kind.
Are you shure that the problem is on the serverside
at all, and not the browser ?
The only thing on the serverside that could cause
something like that would be a timeout in the servlet
(e.g connection or read time out while reading
from a
If all processors are busy, the acceptcount defines the length of
the queue for requests that are waiting for the next free processor.
-Original Message-
From: Chakravarthy, Sundar [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 26, 2003 6:16 PM
To: Tomcat Users List
Subject:
One reason is this:
After a redirect the servlet that issues the redirect will
continue to run unless you stop the processing with a return
statement directly after the redirect. Now consider this
example:
Servlet A:
doSomething();
include(Servlet B);
doSomethingMore();
Servlet B:
Have a look at RUNNING.txt in the tomcat 4 distribution:
(4) Advanced Configuration - Multiple Tomcat 4 Instances
...
(The exact text and number may vary in your distribution,
my RUNNING.txt is quite old)
The following is for unix and tomcat standalone. But it
should be a good starting
I prefer a different setup.
We separate the tomcat installation from the site.
With this approach the conf files are placed outside of
the tomcat directory tree. As long as the config files
are compatible, an upgrade is just a matter of setting a
different CATALINA_HOME.
Althoug we didn't use
Try to strace the tomcat process to find out where
tomcat is looking for the library.
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Tuesday, July 15, 2003 2:53 PM
To: Tomcat Users List
Subject: Re: AW: configuration problem
both library's are
strace is a standard linux program that prints out system calls.
(Like truss under solaris)
You can either attach it to a running process
or trace a program from scratch.
Sources for information:
man strace
http://www.wi.leidenuniv.nl/~wichert/strace/
If you go this way you you have to be carefull with the
alias directive. (If you let tomcat serve everything,
you don't need the alias, but that's quite uncommon)
If you have static resources that are directly delivered
through apache, apache has to alias just this resources,
not the the ones
I don't see any need for this.
You can run a java service on ports 1024 without being root with
portmappers, proxies, iptables and several other tools, that let
sysadmins open well defined holes without compromising the security.
In the long term I would like to see ACL's in the os, that
No it'sdeprecated and not any longer supported.
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Monday, July 21, 2003 10:40 AM
To: [EMAIL PROTECTED]
Subject: for my mod_jk : WARP vs AJP13 ???
Are there any reasons to prefer WARP over AJP ???
What makes you so shure that you have a port problem ?
I think that you are searching in the wrong direction.
As I understand the log message mod_jk complains about
not getting an answer from tomcat after it successfully
sent a request to tomcat. At this point I cann't imagine
a port problem
I would expect that you have to use link href=/html/main.css ...
to get the file. Otherwise the alias directive won't match.
-Original Message-
From: Astrid Wagner [mailto:[EMAIL PROTECTED]
Sent: Thursday, July 24, 2003 12:35 PM
To: [EMAIL PROTECTED]
Subject: problem finding css
Now your are confusing me.
This is what I believe to be true:
Any request can just be served from one DocumentRoot.
Which DocumentRoot is used, is defined by the way you
configure apache
If the request contains a server name that matches a
given virtual host, the setting for
There are a couple of solutions that allow you to achive that.
Which solution is best, depends on your requirements.
Some solutions that require a wrapper script:
- script that calls rsh by super or sudo or something like that
- script that is owned by accountB and has the setuid bit set
No, this shouldn't be a problem if you setup tomcat correctly.
http://jakarta.apache.org/tomcat/tomcat-4.1-doc/proxy-howto.html
(Note: I havn't tried it on my own)
Most documentation that I found about the configuration of tomcat
for running as non root on port 1024 are missing this point:
That doesn't happen under java.
You have to explicitly free/close all resources
that are more valuable than memory (file handles,
db connections, ...)
The finalizer is called by the garbage collector,
which in turn can run any time or not even at all
(if you don't consume enough memory).
From
Sorry. havn't used mysql or dbcp, so can't help you
with this question.
To get answers from others I would suggest, that you
include more infos about the configuration of the pool
and how you access it. (And which version of the pool
you use)
-Original Message-
From: Hans Wichman
I prefer to use tags/labels to store that information.
This way you can use the query interface of the vcs
to retrieve the changes for a given bug id.
I'm not aware of any open sourece project that let you do
this automagically. There are version control systems and
there are bug/issue/request
build a model like
- each user is on the site for n Minutes
- each user makes n requests
- the mean time between to page requests is n seconds
- each page request creates n http requests
- specify the average size for an http request
- specify the average ratio of request/db access
.
One way to verify this, is to use a packet sniffer
and watch the pakets that are exchanged bewenn server
and browser.
Under linux you can use tcpdump.
http://www.tcpdump.org/
tcpdump has also a windows brother (or sister):
http://windump.polito.it/
Under linux and windows you can use
There is someone from xx.xx.xx.xx trying to use an IIS
vulnerability. If it's realy intranet your admin should
have a look at the offending pc if it is infected by a
virus. (Not shure out of the head if this is nimda, code
red or what else)
This vulnerability is not affecting tomcat.
I think you should also include the JDK (vendor and version).
It's not impossible that this might be a JDK problem.
-Original Message-
From: Jeff Tulley [mailto:[EMAIL PROTECTED]
Sent: Tuesday, August 12, 2003 6:41 PM
To: [EMAIL PROTECTED]
Subject: RE: security hole on windows
You have to synchronize the -- and ++ operations.
Otherwise you will have unpredictable results.
You have to keep in mind that activeSessions++
is not atomic, so another thread can get between
the computation of the value and the assignment.
One scenario:
Thread A: read activeSessions = 0
Thread
in this case
synchronized public void sessionCreated(
synchronized public void sessionDestroyed(
should be OK.
-Original Message-
From: Christian Hauser [mailto:[EMAIL PROTECTED]
Sent: Tuesday, August 12, 2003 10:44 AM
To: Tomcat Users List
Subject: Re: HttpSessionListener:
You don't need the useBean.
A bean with session scope doesn't make sense at all,
if you want to access the methods through an object
you can make the object application global.
Or you can make sessionCreated/sessionDestroyed
static and just call them through the class.
-Original
- Each port that is in use must be unique across the
instances of tomcat. So you have to change all ports
that are used. (So the first task, is to throw out
every thing you don't need)
To be more precise each combination of port and IP
must be unique. It's possible to setup differenzt
Can you arrange your file layout in a way, that the jsp's aren't
under the document root for apache ? (I guess they are, otherwise
apache couldn't show them)
-Original Message-
From: Angus Mezick [mailto:[EMAIL PROTECTED]
Sent: Wednesday, August 13, 2003 5:34 PM
To: Tomcat Users
Only if you call encodeUrl on the link.
If you just write out some text it stays as it is
no matter if it contains links.
-Original Message-
From: Shapira, Yoav [mailto:[EMAIL PROTECTED]
Sent: Friday, August 15, 2003 3:35 PM
To: Tomcat Users List
Subject: RE: URLEncoding urls
That is how it works. AFAIK there is nothing you can
configure to change this.
This is a thing that requires some logic to implement a
generic solution. It would require a complete html parser
that parses each response (Quite challanging and time
consuming). And what should this solution do
Are you shure that the browser used https when this error happened ?
This kind of error is also likely to happen, when somebody requests
http:url:443
-Original Message-
From: Schadler Johann [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 12, 2003 1:33 PM
To: [EMAIL PROTECTED]
A Connector consists of two parts:
- A apache module in C (in this case mod_jk2)
- A Tomcat module in java
As the apache module runs as an integral part of
apache, it makes sense to assume that apache is
there to build this part of the connector. The
connector is built against the headers and
Also look for the other ports that may be used by tomcat like 8005/8443/8009.
(Look for all port statements in server xml)
-Original Message-
From: Goehring, Chuck Mr., RCI - San Diego
[mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 19, 2003 9:40 PM
To: Tomcat Users List; Christian
That's not true.
If mod_jk is setup right all requests for the same session
should go to the same tomcat. So you don't need session
replication for this.
One typical reason that this might not happen is to ommit the
jvmRoute=instancname in the engine element of server.xml.
It's also a good
Whenever you use a class without a package name and without
an import it is not a class without a package but a class
that is in the same package as the class that contains the
usage. Since the generated code has a package statement
it would search TemplateDesc in that package. (Where it
It sounds as if the tomcat part of the connector is not listening on port 8009.
Which connector do you use (AjpConnector, or CoyoteConnector with JkCoyoteHandler) ?
-Original Message-
From: Michele Neylon :: Blacknight Solutions
[mailto:[EMAIL PROTECTED]
Sent: Sunday, March 23, 2003
Do you have a context with path= ? Otherwise you will not
be able to access Domain/test.jsp (You would have to insert
the context path: Domain/ContextPath/test.jsp)
test.jsp must be in the top level directory that is given in
the docBase attribute of the context.
-Original Message-
They use many classes that are quite uncommon in 'pure'
servlets.
-Original Message-
From: Collins, Jim [mailto:[EMAIL PROTECTED]
Sent: Monday, March 24, 2003 2:44 PM
To: 'Tomcat Users List'
Subject: RE: Frequent hanging
I can't see what difference a JSP would make because
The context is defined in server.xml.
Have a look for context in this file.
(Has nothing to do with mod_jk* at all)
For specific questions on mod_jk2 I'm the wrong guy,
as I've not used it by now.
-Original Message-
From: Michele Neylon :: Blacknight Solutions
[mailto:[EMAIL
First of all you shouldn't rethrow the exception.
That hides the true nature of the problem.
-Original Message-
From: jsp [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 25, 2003 4:17 AM
To: 'Tomcat Users List'
Subject: MYSQL and TOMCAT
PreparedStatement updateInfo;
That looks more like a problem of the VM or the underlying os
than of tomcat.
What JDK do you use ?
What OS do you use ?
-Original Message-
From: Georges Roux [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 26, 2003 10:53 AM
To: Tomcat Users List
Subject: Unexpected Signal : 11
You might also have a look at
http://www.owasp.org/
That site has some information on common security errors in
webapplications and also some testtools that you can use to
check your site.
-Original Message-
From: Henning Heil [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 27,
That is not possible.
One tomcat instance can only have one vm and java home.
-Original Message-
From: Surendra Kumar [mailto:[EMAIL PROTECTED]
Sent: Monday, March 31, 2003 1:16 PM
To: Tomcat Users List
Subject: Re: Multiple jre's in single tomcat server
What should be the
AFAIK you have to implement it on your own
if the datasource is not a file.
To see how to implement it you might have a look at
the tomcat source:
Although I havn't tried it, I guess yes.
I think you have to define your own RequestWrapper
that lets you set the principal.
-Original Message-
From: Oliver Wulff [mailto:[EMAIL PROTECTED]
Sent: Tuesday, April 01, 2003 2:42 PM
To: [EMAIL PROTECTED]
Subject: user principal, realm
Is it possible that you have a tools.jar in the tomcat classpath
that is older than j2sdk1.4.1_01 ? (AFAIK some versions of
tomcat where delivered with tools.jar)
-Original Message-
From: Robert Priest [mailto:[EMAIL PROTECTED]
Sent: Tuesday, April 01, 2003 10:22 PM
To: [EMAIL
Which class/methods are you talking about ?
-Original Message-
From: Oliver Wulff [mailto:[EMAIL PROTECTED]
Sent: Tuesday, April 01, 2003 5:02 PM
To: Tomcat Users List
Subject: Antwort: RE: user principal, realm
I took a look to JavaDoc and saw that all methods are deprecated.
As I understood your problem I think that you have to
implement your own RequestWrapper as your username is
not in an official attribute but a private one. (So
tomcat wouldn't know how to give this information to
a realm)
More on extending a RequestWrapper:
From the code in
catalina/src/share/org/apache/catalina/authenticator/BasicAuthenticator.java
Principal principal =
((HttpServletRequest) request.getRequest()).getUserPrincipal();
if (principal != null) {
if (debug = 1)
log(Already
I think that the waay of TC406 is better. This way you can
have both solutions.
If you have threads that do something important that has
to be finished before termination, just don't define them
as daemon threads.
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
It's a pretty common solution, you may also look for a cached ResultSet:
http://freshmeat.net/projects/crs/?topic_id=66
http://developer.java.sun.com/developer/earlyAccess/crs/ (Requires login)
http://www.javaworld.com/javaworld/jw-02-2001/jw-0202-cachedrow.html
Some of our reasons to do this:
better separation between the applications:
- If one webbapp needs to go down for maintainance, the others can
still run.
- Better control on the resource usage.
As you can't define memory settings on the context level, there
is no way to retrict
There are many possible reasons why tomcat may freeze:
- Database
- ConnectionPool exhausted
Do you use a connection pool ?
- Which one ? How is it configured ?
- Most connection pools let you configure the behaviour
if the the pool is exhausted. To identify the problem
In this case I don't completly agree with you.
Through the implementation of a servlet container you
can incluence the behaviour of the gc. It depends on how
many objects are created by the different containers
to complete the same task and how the lifecycle of the
created objects is. Given
Looks like one of your xml files has a syntax error.
The error message just doesn't indicate which one.
So you have to look around (in line 15) in
server.xml
web.xml
*.xml
-Original Message-
From: Chris Daly [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 09,
You don't need apache and iis.
It's either apache or iis (or tomcat stand alone)
but normaly not both of them. (Although this
my not be impossible, just very unlikely)
-Original Message-
From: LILES, DAVID (CONTRACTOR) [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 09, 2004 2:59 PM
I guess they where mainly for the leaks in the web applications
not because they where thinking that they had leaks in the pool.
-Original Message-
From: Hofmann, Benjamin [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 09, 2004 4:20 PM
To: '[EMAIL PROTECTED]'
Subject: JNDI
Have you tried to search for
windows+task+manager+memory
I just get 30 result.
(If there is one that satisfies you is a different question)
-Original Message-
From: Allistair Crossley [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 10, 2004 12:06 PM
To: [EMAIL PROTECTED]
Cc: SH
Which jdk do you use ?
There was once a version that had an error in ready().
As far as I can remember it always returned true.
Try to end the loop also when you couldn't read a line.
(Depends on what getLogentry() does in this case)
-Original Message-
From: Adam Buglass
Scenario 1 will use more memory than scenarion 2.
The ratio depends on the internal architecture of the classes.
(What kind of object they create, where they store it)
If you just look at the core size that is needed to load
the classes it roughly directly proportional to the number
of webapp
Although I think that the solution is correct, (I have given
the same advice) I think that the original solution should
work (according to the apidoc):
From BuffereReader.ready():
Tell whether this stream is ready to be read. A buffered
character stream is ready if the buffer is not empty, or
Deny takes precedence over allow.
(The valve is missing the option to define the order)
I'm not shure if that explain all your problems but some.
http://cvs.apache.org/viewcvs.cgi/jakarta-tomcat-catalina/catalina/src/share/org/apache/catalina/valves/RequestFilterValve.java?rev=1.3view=auto
Do you have dnslookups enabled/disabled ?
If this is disabled the hostname ist the IP.
-Original Message-
From: Jason Keltz [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 10, 2004 4:43 PM
To: Tomcat Users List
Subject: RE: Restrict to specific IP's
However, something
Your problem doesn't have to do much with the 'most stable tamcat'.
This is an jvm error. (Although a change in the tomcat version may
get you around this bug, but that is accidently)
- Make shure you have all required os patches for your os.
- Maybe you have to look for a newer jvm version for
For the os patches have a look at:
http://sunsolve.sun.com/pub-cgi/show.pl?target=patches/J2SE
-Original Message-
From: Ralph Einfeldt
Sent: Thursday, March 11, 2004 7:39 AM
To: Tomcat Users List
Subject: RE: Stable Version
Your problem doesn't have to do much with the 'most
According to the spec (Chapter 11.2):
spec
A string beginning with a / character and ending
with a /* postfix is used for path mapping.
A string beginning with a *. prefix is used as an
extension mapping.
A string containing only the / character indicates
the default servlet of the
A redirect creates a new request.
If you want to pass informations across a redirect
you can use url parameters in the redirect url.
Maybe it's is an option to you to forward to the jsp,
then no new request is created.
-Original Message-
From: Edson Alves Pereira [mailto:[EMAIL
That has been answered 40 Minutes ago:
A redirect creates a new request.
If you want to pass informations across a redirect
you can use url parameters in the redirect url.
Maybe it's is an option to you to forward to the jsp,
then no new request is created.
-Original Message-
From:
That is the fourth mail with the same question.
2 times it has been answered.
Didn't you like the answer, didn't you receive it ?
-Original Message-
From: Edson Alves Pereira [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 11, 2004 8:06 PM
To: 'Tomcat-User List'
Subject: doubts
One tomcat tool is the RequestDumperValve.
http://jakarta.apache.org/tomcat/tomcat-4.1-doc/config/valve.html
(The doc is a bit sparse, so it needs some testing
what is logged and if that is enough)
-Original Message-
From: Adam Hardy [mailto:[EMAIL PROTECTED]
Sent: Friday, March 12,
I don't think that this message has anything to with the lost
connection pool.
(Do you really mean 'connection pool' and not just 'connection' ?)
If you lose your connection, it may be caused by the databaseserver
that cancels idle connections after a while.
Look for validationQuery in:
I don't see any reason why an object in the context
should have any thing to do with the stacktrace.
What makes you think that they are related ?
What makes you think that the pool is lost ?
(context.getAttribute(AttribName) == null ?)
Or do you just don't get connections from the pool ?
Or
Tomcat doesn't require a certain jvm (as long as it has the right vdk version).
Do you start tomcat as a service ?
-Original Message-
From: stanley [mailto:[EMAIL PROTECTED]
Sent: Monday, March 15, 2004 2:44 PM
To: [EMAIL PROTECTED]
Subject: Tomcat5 can not use JRockit
Hi
- Original Message -
From: Ralph Einfeldt [EMAIL PROTECTED]
To: Tomcat Users List [EMAIL PROTECTED]
Sent: Monday, March 15, 2004 9:48 PM
Subject: RE: Tomcat5 can not use JRockit
Tomcat doesn't require a certain jvm (as long as it has the
right vdk version
I just stumbled across the following link:
http://www.modsecurity.org/products/modsecurity/java/index.html
If somebody thinks it's interesting [s]he can keep it :),
otherwise forget it :{.
-
To unsubscribe, e-mail: [EMAIL
That is not a matter of tomcat but of the jvm and the os.
If the jvm uses more than one cpu on this os tomcat will
do the same.
- What jvm are you using (vendor / version) ?
- Which jvm options do you set ?
-Original Message-
From: John Hilton - CPX COA [mailto:[EMAIL PROTECTED]
That may be a bit to simple, as this is just performing one
request after the other. You may have to force concurrent
request:
i=1
while [ $i -lt 2 ] ; do
j=1
while [ $j -lt 10 ] ; do
# start lynx in background
lynx -source $URL /dev/null
j=`expr $j + 1`
done
wait #
To your question: I don't know.
Having a delay in the jsp (or servlet) isn't enough to
enshure concurrent requests. You have to do something
on the calling site to enforce this.
If you just call lynx as in your example, lynx will
wait until the response is there. So if you delay the
jsp, you
901 - 1000 of 1242 matches
Mail list logo
