BUG? tomcat & Solaris8 on Sparc

Dear all,
we have encounterd a problem we like to mark as a bug in Solaris8 and or in all versions of Apache Tomcat.
What is te problem and how did we find out?
We upgraded from Solaris7 to Solaris8 a couple of days ago. We started te server and let it run fore a couple of days in test enviroment. No problems to reporte so we switch the Solaris7 and the solaris8 server and where runninf fine the Solaris8 machine as a production server.
After a setting change of Tomcat we suddenly encounterd a lot of crashes from Tomcat whitout any reason. After a day of searing on the web could not find a solution so we put the old version of Tomcat back to the Solaris8 server in the hope that the problem was solved. Then things where getting clear.... Tomcat was still crashing.
What is the problem.
On solaris8 Tomcat is unstabele when you start it from a telnet session. After you kill the telnet session tomcat is trying to write messages to the telnet session and is crashing when that session is gone.
What is the solution.
The solution is simpel use the NOHUP option to start tomcat. like this:
nohup ./tomcat.sh start
Why we think it is a bug.
- The reason we think it is a bug is that this problem was never encounterd in Solaris7.
- A second reason is that we not put new servlets so te bug is not in the software we have written, this software is already running fore a year on Solaris7.
- We think it is a bug in Solaris because the problem is comming up with every version of Tomcat that we tested running on Solaris8.
Are there more people that have encounterd this problem and are there people how think this is not a bug?
Best regards,
Johan Louwers.

This is definitely **NOT** a Solaris bug.
Most likely you've been trapped by a change to the shell that handles SIGHUP protection different in Solaris 8 than in Solaris 7.
Generally, if you run a program in the backgound that does read from stdin and/or write to stdout/stderr you **MUST** redirect stdin/stdout/stderr to some file or /dev/null to avoid funnies and/or screen corruption. Additionally you must protect it from dying on SIGHUP reception if it does not handle it itself. This is what 'nohup' is for. Otherwise your program might die if you exit your shell and/or logout. 'nohup' automagically does i/o redirection for you, BTW.
- What shell are you using?
- SUN people: any comments?

Similar Messages

Has anyone installed 9i for Solaris8(for SPARC) on Solaris8- X86 (for Intel)

Has anyone installed 9i (9.0.1.1) on solaris8-X86 (intel). I know it is not available on the Oracle website for download. I was wondering if anyone has tried installing the sparc version on Intel.
I have done the same for 8i but don't know if it can be done for 9i.
Thanks

I know this seems silly, but an "eval" version of 9i for Solaris Intel was available from the Oracle Store for the cost of the media. I have also found when I went from Solaris 8 10/01 to Solaris 8 02/02 that 9i came with the OS for the Sparc. I haven't upgraded my Intel boxes yet so I don't know if the same trick applies!
Not quite what you asked, but....
Colin.

Solaris x86, apache2-mod_jk and SSL

I have run into a strange problem.
I have installed apache 2.2.11 from source. Installed nTrust certs. Everything works fine.
I also installed Tomcat 5.5.27. Tested by itself port 8080 etc. all works fine.
Then, mod_jk compiled the latest 1.2.28 from source (no binaries available).
Set up workers.properties etc.
worker.list=worker1
worker.worker1.type=ajp13
worker.worker1.host=localhost
worker.worker1.port=8009
httpd.conf
JkWorkersFile /usr/local/apache2/conf/workers.properties
JkShmFile /usr/local/apache2/logs/mod_jk.shm
JkLogFile /usr/local/apache2/logs/mod_jk.log
JkLogLevel info
JkLogStampFormat "[%a %b %d %H:%M:%S %Y] "
jkMount /* worker1
So, I am forwarding everything to Tomcat
http://<server>/ comes back with the usual Tomcat starter page
But, go SSL
https://<server>/ comes back with Apache "It works!"
Looks like mod_jk is not getting it !
Perhaps, it's a config issue.
But, similar config worked fine with older versions of Apache/Tomcat/mod_jk on Sparc!!!
Any help would be appreciated.
Thanks

We're having the same problem, although with CF11.

Solaris 8 patches

Hi could not find a better forum to post this...
I have a Sun Fire server running Solaris8 on Sparc. We had some security issues with this server and I was going to download the patch set's as we have done before.. but I get the happy message "This download is only available to users with a valid Sun Service Plan associated to their SunSolve user account".
Now do I really need to purchase ANYTHING to get OS patches? I paid for the OS and the server... Even Micro$soft gives away patches for free.. I don't get this.. I assume there must be a way?
I don't really mind paying... but for patches?
Any tips? Seems abit daft paying a for a service agreement on an old server with an old os.. we are stuck with some other old software so we cannot upgrade this server to solaris 9 or 10... actually we even downgraded it when we got it..
K

I just took a moment
(admittdly a very brief 90-second moment)
and went to the Sun corporate web site.
In the top right corner of that web page there is a search box.
I searched on "smpatch" and got more than 600 results, including links to documentation.
I did not spend too much time reading the contents of the links,
but it is probable that smpatch might be only for Solaris 9 and newer.
... but I don't have any Solaris 8 systems to experiment with.

Anyone running Communications Suite 5 on X86 hardware

Hi,
We recently installed communications Suite 5 on X86 hardware and have run into numerous bug and issues. We are wondering if there are others that have installed Comm. Suite 5 on X86 that are willing to share their experiences and possibly collaborate with us.
We've contacted support many times and on many occasions we get bug reports and patches that refer to bug fixes for Sparc Hardware but it doesn't seem that many people (including Sun support) use x86.
Any information or comments would be appreciated.
thanks

Hi Shane,
I appreciate the response.
We are as patched as we can. We have also been putting in binary files, IDRs and T patches as recommended by our support person(s), and many times having to back them out because the patch caused as many problems as it has fixed. The only time the system was at all stable and working to what we thought was proper was during our (as we later found out, minimal) testing. Once we actually had most of our user population using it and putting it under load we've had problem after problem. Including things such as the Calendar server crashing ever 1-2 hours, Calendar database getting corrupted and then deleted by failing hot backups/archiving, mal-formatted email message, ics* attributes being deleted for users when setting permission via outlook connector, uwc, or cal express, and numerous JavaScript errors.
Maybe the Sparc version is just as buggy but I don't think that it is a far stretch to think that completely different hardware platforms would not always have the same bugs. Especially considering that we are talking completely different architectures with different endian-ness, memory registers, etc...
I'm sure a vast majority of bugs in jsp, java, and javascript code can be seen between platforms but as noted above we have been given binaries which, maybe have the same code base but wont necessarily cause the same bugs. Maybe the bugs are there but on many occasions I have been given a bug number which I can't pull any information about (being that it is an internal bug number I suppose). So maybe the bugs are in both sparc and x86 but I have no way to know this until the official patch is release or the bug is publicly available.
I posted here, so that I could get a better idea of what others are seeing with their systems. Another reason I ask this and stated that it seemed like very few people are using comm suite 5 on x86, because from the start of our testing and throughout the last 5 months we have repeatedly asked our Sun partner, Sun support personel, Sun account managers, and anyone else involved, if they could refer us to someone that has a comm suite 5 deployment on x86 and as yet we have not gotten a single response.
With this said, I am simply stating my experiences up until now and on many occassions (of the nearly 40 cases that we have opened with support) I have repeatedly been told things like my only test platform is a sparc ultra 5, or we can't recreate your issue on a similar system (similar as in sparc and not x86 or deployed on a single host and not a DWP setup like ours). I've created test accounts for support to be able to log into our uwc so that they can see what we are seeing.
I have always thought very highly of Sun but this has been changing slowly over the last 5 months or so.
You asked for bug numbers, here are some, maybe I am wrong that they are particular to x86, maybe it is because we have a DWP setup, I've been wrong before.
6590680,6591407
Again, I do appreciate any response, and I would like to say thanks to all the support persons that we have been dealing with to get our issues resolved.
As another aside, it might be helpful to know what our setup might be like so here it is:
We have three servers 2 Sun Fire x2100's, one Sun Fire x4200, a Sun Stroragetek 6140 Disk Array and a Sun Storedge C4 Tape Library.
x2100 is used for front end calendar server, UWC, calendar express, messenger express (users don't go directly to messenger express, but we have used calendar express for some workarounds temporarily and most users don't use or know about either calendar or messenger express, they just use UWC).
x2100 for MMP and MTA, this is connected via LMTP to our backend Message store.
x4100 for Messages Store and back end Calendar Server, our x4100 also is running a Solaris Zone where we have our Directory server (this zone in on zfs as was recommended by our Sun partner, until recently, I'm told this was a bad idea)). The x4100 message store (zfs), calendar database (zfs), and mboxlist db (zfs) is stored on the 6140 disk array. The x4100 is connected to the 6140 via 2 HBA's through a Qlogic Fibre channel switch.
thanks again and sorry for the blogish style post.

Steps to configure IP Address

Can any one explain me the full steps to change ip address in solaris 2.5,7 & solaris8 on sparc & x86 platform?
any differences are in commands in all versions? if yes explain me

Perhaps:
man sys-unconfig
A Google search found
http://www.scit.wlv.ac.uk/cgi-bin/mansec?1M+sys-unconfig

Chinese encoding problem under Mozilla or Netscape 7

We have a Web site with Chinese content in big5 character set. There are two
kinds of content - static content which are in html format and dynamic
content which are JSP/Servlet hosted by WebLogic.
We've put the following META tag to all of our Chinese content :
<meta http-equiv="Content-Type" content="text/html; charset=big5">
However, we found that under all variants of Mozilla browser (Netscape 7,
Mozilla, etc.), only the static Chinese content and dynamic content in
Servlet will be displayed with big5 encoding. For dynamic content in JSP, we
have to manually switch the encoding to big5.
This problem does not occur in IE, nor Netscape 4.x.
I've checked many other Web sites and it seems that Web sites in UTF8 may
not have this problem, but I'm not sure.
Any idea ?
p.s. Our Web site is http://www.dahsing.com.

hi,
I do not thing that the problem is Windows. I have the same problem on a linux server, config :
Red Hat 7.3
JWSDP1.3
j2ee 1.4.2
but the problem appeared when I changed JDK from 1.4.1 to 1.4.2.
So, I can see two possible issues :
- jdk 1.4.2 has a bug
- tomcat server is not compatible with this version

Problem with Preprocessor Compilation.

Hi,
I am working with Forte C++ Personal Ed 6/SPARC on Solaris8 on SPARC platform. I am intrested in generating foo.i file from foo.idl using below command.
CC -P -I/export/narendra/project/idl foo.idl
This should produce foo.i file in /export/narendra/project/idl directory.
Instead I get a message " Could not open input file foo.idl.
Can anyone help me out in this regard.
Narendra Kumar.
[email protected]

Hi,
Can you check and make sure whether you are compiling in the directory where you have "foo.idl" ?
You can try specifying the full path as:
CC -P -I/export/narendra/project/idl /export/narendra/project/foo.idl
I tried compiling a sample program with Forte 6 Update 1 and it
works fine.
Thanks,
Gopinath.

Solaris8 and 9 (possibly 7) /dev/poll driver bug report.

Hello,
I'd like to report a bug in the solaris 8 and 9 /dev/poll driver (poll(7d)).
As i do not have a support account with sun or anything like that, there
seems to be no other way to do that here (which is of course a very sad
thing).
Bug details:
The /dev/poll device provides an ioctl-request (DP_ISPOLLED) for checking
if a particular filedescriptor is currently in the set of monitored
filedescriptors for that particular /dev/poll fd set (open /dev/poll fd).
A quote from the documentation of the poll(7d) manual page taken from
Solaris9:
"DP_ISPOLLED ioctl allows you to query if a file descriptor is already in
the monitored set represented by fd. The fd field of the pollfd structure
indicates the file descriptor of interest. The DP_ISPOLLED ioctl returns 1
if the file descriptor is in the set. The events field contains the
currently polled events. The revents field contains 0. The ioctl returns 0
if the file descriptor is not in the set. The pollfd structure pointed by
pfd is not modified. The ioctl returns a -1 if the call fails."
It says that when you query for an filedescriptor which is currently being
monitored in the set, that it would return 1, and change the events field of
the pollfd structure to the events it's currently monitoring that fd for.
The revents field would be set to zero.
However the only thing which actually happens here, is that FD_ISPOLLED
returns 1 when the fd is in the set and 0 if not. When the fd is in the
set, when FD_ISPOLLED returns 1, the events field remains unmodified, but
the revents field gets changed.
A small sample code to illustrate:
#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/devpoll.h>
main() {
struct pollfd a;
int dp_fd = open("/dev/poll", O_WRONLY);
a.fd = 0; /* stdin */
a.events = POLLIN; /* we monitor for readability, POLLIN=1 */
a.revents = 0;
write(dp_fd, &a, sizeof(a));
a.fd = 0;
a.events = 34; /* filled in with bogus number to show malfunctioning */
a.revents = 0;
printf("DP_ISPOLLED returns: %d\n", ioctl(dp_fd, DP_ISPOLLED, &a));
printf("a.fd=%d, a.events=%hd, a.revents=%hd\n", a.fd, a.events,
a.revents);
According to the documentation of /dev/poll and namely DP_ISPOLLED this
program is supposed to print the following:
DP_ISPOLLED returns: 1
a.fd=0, a.events=1, a.revents=0
However it prints the following:
DP_ISPOLLED returns: 1
a.fd=0, a.events=34, a.revents=1
You can take any number instead of '34' and it will simply remain untouched
after the DP_ISPOLLED ioctl-request.
I hope it's clear now that the solaris8 and solaris9 (and probably solaris7
with /dev/poll patch too) DP_ISPOLLED implementation is broken.
This bug is also easily illustrated by looking at the solaris8 kernel sourcecode:
<snippet osnet_volume/usr/src/uts/common/io/devpoll.c:dpioctl()>
case DP_ISPOLLED:
pollfd_t pollfd;
polldat_t *pdp;
if (pollfd.fd < 0) {
mutex_exit(&pcp->pc_lock);
break;
pdp = pcache_lookup_fd(pcp, pollfd.fd);
if ((pdp != NULL) && (pdp->pd_fd == pollfd.fd) &&
(pdp->pd_fp != NULL)) {
pollfd.revents = pdp->pd_events;
if (copyout(&pollfd, (caddr_t)arg,
sizeof(pollfd_t))) {
mutex_exit(&pcp->pc_lock);
DP_REFRELE(dpep);
return (set_errno(EFAULT));
*rvalp = 1;
</snippet>
its' clearly visible that the code writes the current monitored events to
the revents field:
'pollfd.revents = pdp->pd_events;'
and that it doesnt set revents to zero.
It's funny to see that this has been like this since Solaris8 (possibly 7). That means nobody ever used DP_ISPOLLED that way or people were simply to lazy to file a bug report.
Another funny thing related to this. is that Hewlett-Packard did seem to know about this. Since HP-UX11i version 1.6 they also support /dev/poll. From their manual page i ll quote some sentences from their WARNING session:
"The ioctl(DP_ISPOLLED) system call also returns its result in the revents member of the pollfd structure, in order to be compatible with the implementation of the /dev/poll driver by some other vendors."
Hopefully this will get fixed.
I also like to reexpress my very negative feelings towards the fact that you're not able to file bug reports when you do not have a support contract. Ridiculous.
Thanks,
bighawk

Have I mentioned how much i love my playbook now Great job on os 2.0

Tomcat 5.5.12 with Solaris 10 on Sparc SunFire T2000

I am experiencing difficulty with installing Tomcat 5.5.12 on a Solaris 10 Sparc SunFire T2000 SunOS machine. Can someone please help?
After I untar the apache-tomcat-5.5.12.tar.gz like this:
gunzip < apache-tomcat-5.5.12.tar.gz | tar xvf -
I then tried to start the Tomcat like this:
[root@sun] ./bin/startup.sh
Using CATALINA_BASE: /opt/tomcat/apache-tomcat-5.5.12
Using CATALINA_HOME: /opt/tomcat/apache-tomcat-5.5.12
Using CATALINA_TMPDIR: /opt/tomcat/apache-tomcat-5.5.12/temp
Using JRE_HOME: /usr/bin
[root@sun]
But the Tomcat is not up, and the log file only says:
This release of Apache Tomcat was packaged to run on J2SE 5.0
or later. It can be run on earlier JVMs by downloading and
installing a compatibility package from the Apache Tomcat
binary download page.
I have searched on the Internet and found some similar info that talks about the five steps to install Tomcat on a Solaris (http://johnturner.com/howto/apache2-tomcat4112-sol8-howto.html):
1. Build/Install Apache
2. Install JDK
3. Build/Install Tomcat
4. Build/Install JK Connector (between Apache and Tomcat)
5. Configure the whole setup and test both Apache and Tomcat.
However, I was also told that I should be able to just unzip the Tomcat and startup the sample web page.
Do I have to go through the whole five major steps to bring up Tomcat 5.5.12 on Solaris 10? I know that I need to use some GNU tools instead of Solaris tool such as tar for this effort. I'd like to avoid the major effort if all I need to just Tomcat.
I'd appreciate any help that you may have.
- Andy

I guess tomcat 5.5.20 be installed on JDK 1.4 using the comaptiblity package given in link http://tomcat.apache.org/download-55.cgi#5.5.20.
But I have not tried.
Does any one know how to ?
Thanks in advance.

Tomcat 4 bug in tomcat 5

hi,
What i have:
- eclipse Version: 3.0.0
Build id: 200406251208
- lomboz 3.0.0
- tomcat 5
what i want to do is:
i want to enable jsp-debugging in eclipse using lomboz, so i followed the steps on the objectlearn-website to achieve this.
i changed the context path so every generated servlet goes into the /j2src folder and i can set breakpoints, which is fine, but now i got a problem which should not occur with tomcat 5 because it was a bug in tomcat 4.
------ quote from the sysdeo website ------
Known problem 1 : Tomcat 4 and JSP in project subdirectories
Generated servlets for JSP couldn't be compile by Eclipse. Subdirectories do not appear in Tomcat 4 generates servlets. Package definition is always package org.apache.jsp.
----- end quote ------------
thats exactly the problem i got right now, package definition is always org.apache.jsp, but there is no patch for this problem for tomcat 5, it isnt even mentioned on the website for tomcat 5,
so what is wrong and how can i change this?
thanks in advance, timo

is there nobody here familiar with lomboz and tomcat 5???
if there is somebody in this forum who uses tomcat 5 and lomboz:
did you get the same error like me, i.e. that lomboz automatically throws all generated servlets into the package org.apache.jsp??
if the answer is yes, what did you do to fix this bug, there is no patch for tomcat 5 for this bug, since this bug doesnt seem to exsist officially in tomcat 5 ?!?
if the answer is no, what tomcat-lomboz-environment did you set up? did you follow the guidelines on the objectlearn-website for activating jsp-debugging or did you do something else / more?
damn, i posted this problem in a few forums and i seem to be the only person on the planet having this problem or using this combination...

Tomcat 6 failing to evaluate EL expressions, Tomcat 6 bug?!!!!

Can someone please tell me this is not a bug in Tomcat 6, its too obvious and to some extend unacceptable I must be doing something wrong.
Here is my problem, I have written my own custom component to display a div tag.
the JSP is as follows:
<faces:view>
<html>
    <head>
        <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
        <title>My Custom Tags</title>
    </head>
    <body>
      <html:form>
      <h2>Cat In The Rain</h2>
      <mine:mytag title="#{index.title}">Dancing in the rain.</mine:mytag>
      </html:form>
    </body>
</html>
</faces:view>The tag class is given below:
This text #{index.title} is being interpreted as Literal text which is obviously wrong, when deployed in glassfish the code works fine.
Output text in my tomcat console is
MyTag.setProperties(): Title is literal text....
MyTag.setProperties(): title value= #{index.title}
MyTag.setProperties(): Title Value expression is null
The Tag class.
public class MyTag extends UIComponentELTag{
private ValueExpression title;
public String getComponentType() {
    return "test.MyOutput";
public String getRendererType() {
    return "test.MyRenderer";
@Override()
protected void setProperties(UIComponent component) {
    super.setProperties(component);
    if (null!= title){
      if (!title.isLiteralText()){
        component.setValueExpression("component.title", title);
        System.out.println("MyTag.setProperties(): Its not literal text.");
        System.out.println("MyTag.setProperties ():              title value = "+ title.getValue(getELContext()));
      else{
        component.getAttributes().put("component.title", title.getExpressionString());
        System.out.println ("MyTag.setProperties(): Title is literal text....");
        System.out.println("MyTag.setProperties():              title value= "+ title.getExpressionString());
    System.out.println ("MyTag.setProperties(): Title Value expression is "+ component.getValueExpression("component.title"));
}The component class.
public class MyOutput extends UICommand{
public MyOutput(){
    setRendererType("test.MyRenderer");
@Override()
public String getFamily() {
    return "test.MyOutput";
}The tld file
<?xml version=" 1.0 " encoding="UTF-8"?>
<taglib version="2.0"
        xmlns="http://java.sun.com/xml/ns/j2ee "
        xmlns:xsi=" http://www.w3.org/2001/XMLSchema-instance"
        xsi:schemaLocation="http://java.sun.com/xml/ns/j2ee web-jsptaglibrary_2_0.xsd">
<tlib-version> 1.0</tlib-version>
<short-name>mine</short-name>
<uri>http://www.my-tests.com </uri>
<tag>
    <name>mytag</name>
    <tag-class> test.MyTag</tag-class>
    <body-content>JSP</body
-content>
    <description></description>
    <attribute>
      <name>binding</name>
      <required>false</required>
      <rtexprvalue>true</rtexprvalue>
      <deferred-value>
        <type>test.MyOutput</type>
      </deferred-value>
      <description>
          A ValueExpression that resolves to the UIComponent that corresponds
          to this tag. This binding allows the Java bean that contains the UIComponent
          to manipulate the UIComponent, its properties, and its children.
      </description>
    </attribute>
    <attribute>
      <name>title</name>
      <required>false</required>
      <deferred-value>
        <type>java.lang.String</type>
      </deferred-value>
      <description></description>
    </attribute>
        <attribute>
            <name>rendered</name>
            <required>false</required>
            <deferred-value>
                <type>boolean</type>
            </deferred-value>
            <description><![CDATA[ Use the rendered attribute to indicate whether the HTML code for the
component should be included in the rendered HTML page. If set to false,
the rendered HTML page does not include the HTML for the component. If
the component is not rendered, it is also not processed on any subsequent
form submission.
]]></description>
        </attribute>
        <attribute>
            <name>id</name>
            <required>false</required>
            <rtexprvalue>true</rtexprvalue>
        </attribute>
</tag>
</taglib>

Set the JSP version in your TLD to 2.1.
Meaning from:
<taglib version="2.0"
        xmlns="http://java.sun.com/xml/ns/j2ee "
        xmlns:xsi=" http://www.w3.org/2001/XMLSchema-instance"
        xsi:schemaLocation="http://java.sun.com/xml/ns/j2ee web-jsptaglibrary_2_0.xsd">to:
<taglib version="2.1"
        xmlns="http://java.sun.com/xml/ns/j2ee "
        xmlns:xsi=" http://www.w3.org/2001/XMLSchema-instance"
        xsi:schemaLocation="http://java.sun.com/xml/ns/j2ee web-jsptaglibrary_2_0.xsd">Edited by: rlubke on Oct 15, 2007 9:32 PM

[Tomcat bug?] Calling response.sendError(404) from JSP

Calling "response.sendError(404);" from JSP context causes Tomcat to mangle all non-latin characters.
Any ideas why?
How to recreate bug:
http://www.kitoy.ru/int/bug1.zip
There are 3 files in bug1.zip.
index..jsp:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<%@page pageEncoding="utf-8"%>
<%@page contentType="text/html; charset=utf-8"%>
<%
    response.sendError(404);
%>
<html>
</html>404.jsp:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<%@page pageEncoding="utf-8"%>
<%@page contentType="text/html; charset=utf-8"%>
<html>
<h1>Hello in Russian: Привет!</h1>
</html>and WEB-INF\web.xml:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE web-app PUBLIC "-//Sun Microsystems, Inc.//DTD Web Application 2.3//EN" "http://java.sun.com/dtd/web-app_2_3.dtd">
<web-app xmlns="http://java.sun.com/xml/ns/j2ee"
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xsi:schemaLocation="http://java.sun.com/xml/ns/j2ee http://java.sun.com/xml/ns/j2ee/web-app_2_4.xsd"
    version="2.4">
    <display-name>bug1</display-name>
    <description>bug1</description>
    <distributable/>
    <error-page>
        <error-code>404</error-code>
        <location>/404.jsp</location>
    </error-page>
</web-app>Calling of index.jsp should cause 404 error with custom error page. The error page is utf-8 encoded and contains russian characters. That's by design.
In fact, Tomcat shows ?????? instead Russian variant of "Hello".
Tomcat/5.5.17 or Tomcat/5.0.28.- it does not matter.
Is it Tomcat bug or I should not call response.sendError(404) from JSP?

it's not a tomcat bug, it's rather a problem with
your code.
change index.jsp like the following then test
it !
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01
Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<%@page pageEncoding="utf-8"%>
<%@page contentType="text/html; charset=utf-8"%>
<%
esponse.sendRedirect("blablablablablabla.html");
%>
<html>
</html>
Thanks for the insight. But the issue here is with the response.sendError() method.

[svn] 1605: Bug: BLZ-155 - Deadlock using a thread pool for tomcat connectors

Revision: 1605
Author: [email protected]
Date: 2008-05-07 14:24:53 -0700 (Wed, 07 May 2008)
Log Message:
Bug: BLZ-155 - Deadlock using a thread pool for tomcat connectors
QA: Yes - Jorge verified in the QE lab
Doc: No
Ticket Links:
http://bugs.adobe.com/jira/browse/BLZ-155
Modified Paths:
blazeds/branches/3.0.x/modules/core/src/java/flex/messaging/client/EndpointPushNotifier.j ava

Cross post: http://forum.java.sun.com/thread.jspa?threadID=5215686&tstart=0
If you must post across forums, just post in the most relevant one and then give the link to that in the other posts.

[svn] 939: Bug: BLZ-67 - Should turn off session persistence in the Tomcat server we ship.

Revision: 939
Author: [email protected]
Date: 2008-03-26 13:35:44 -0700 (Wed, 26 Mar 2008)
Log Message:
Bug: BLZ-67 - Should turn off session persistence in the Tomcat server we ship.
QA: No
Doc: No
Checkintests: Pass
Details - Turning off session persistence in the version of Tomcat that ships with BlazeDS. We don't store much in the session so session persistence doesn't make much sense for us. Users should feel free to turn this back on if they feel they need it.
Ticket Links:
http://bugs.adobe.com/jira/browse/BLZ-67
Modified Paths:
blazeds/branches/3.0.x/servers/apache-tomcat-6.0.14/conf/context.xml

Hi, thank you for your replies, I found out few things about my servlet, and its portability
and i have few questions, although i marked this topic as answered i guess its ok to post
I am using javax.servlet.context.tempdir to store my files in that servletcontext temporary directory. But i dont know how to give hyperlink
of the modified files to the user for them to download the modified files.
What i am using to get the tempdir i will paste
File baseurl = (File)this.getServletContext().getAttribute("javax.servlet.context.tempdir");
System.out.println(baseurl);
baseurl = new File(baseurl.getAbsolutePath()+File.separator+"temp"+File.separator+"files");
baseurl.mkdirs();so i am storing my files in that temp/files folder and the servlet processes them and modifies them, then how to present them as
links to the user for download ?
and as the servlet is multithreaded by nature, if my servlet gets 2 different requests with same file names, i guess one of them will be overwritten
And i want to create unique directory for each request made to the servlet , so file names dont clash.
one another thing is that i want my servlet to be executed by my <form action> only, I dont want the user to simply type url and trigger the servlet
Reply A.S.A.P. please..
Thanks and regards,
Mihir Pandya

BUG? tomcat & Solaris8 on Sparc

Similar Messages

Maybe you are looking for