Re: RPM Pacakages for RHLinux

[Yu Meng Chong]

Hari Om wrote:

where can I download rpm file for JDK and Tomcat on Red Hat Linux  7.1
I tried rpmfind.net freshrpms.net and rpmseek.com
I use the rpm version of IBM's Java, which you can download from :
https://www6.software.ibm.com/dl/lxdk/lxdk-p
what is better way to INSTALL above pacakages on Red Hat 
Linux...(1)TAR/GZ or (2) RPM...?
They're both the same, more or less. For the tarball, you have to create 
a directory and untar Java into that. For the rpm, you can just install 
using rpm -ivh xxx.

Where is the FORUM for RED HAT LINUX...?
I don't know ! I use Red Hat on my laptop, but am considering shifting 
back to TurboLinux because of double byte support.

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

Re: Re: Unable to create jar cache

[Yu-Meng Chong]

You are absolutely right !!! Looks like 13 is the unlucky number for IBM
(their most recent release is SR13). I reverted back to the January Service
Release of IBM's JDK and the error disappeared. I guess I'll have to
feedback this bug on the IBM newsgroup...

Thanks a million, Tom !! You saved me a LOT of time !!

Regards,
pascal chong


- Original Message -
From: <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, March 21, 2002 7:35 PM
Subject: Odp: Re: Unable to create jar cache


>
> Hi !
>
> I've changed VM from Sun to IBM and now i have exactly the same problem,
so
> maybe it is not tomcat false.
>
> Any other suggestions ?
>
> Tom
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>


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

Re: Re: Unable to create jar cache

[Yu-Meng Chong]

Hey ! You could be on to something here !! I'm using the IBM JDK as well !!
I'm going to check this out right now ...

Regards,
pascal chong


- Original Message -
From: <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, March 21, 2002 7:35 PM
Subject: Odp: Re: Unable to create jar cache


>
> Hi !
>
> I've changed VM from Sun to IBM and now i have exactly the same problem,
so
> maybe it is not tomcat false.
>
> Any other suggestions ?
>
> Tom
>
>
> --
> To unsubscribe:   
> For additional commands: 
> Troubles with the list: 
>
>


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

Re: Unable to create jar cache

[Yu-Meng Chong]

Hmm... well, according to "df" it says that only 4% of the /tmp partition is
being used up.

Thanks and Regards,
pascal chong


- Original Message -
From: "James Williamson" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Thursday, March 21, 2002 6:21 PM
Subject: Re: Unable to create jar cache


> You sure you haven't run out of disk space?
>
> James Williamson
> www.nameonthe.net
> - Original Message -
> From: Yu-Meng Chong <[EMAIL PROTECTED]>
> To: Tomcat Users List <[EMAIL PROTECTED]>
> Sent: Thursday, March 21, 2002 10:28 AM
> Subject: Unable to create jar cache
>
>
> > Hi All !
> >
> > I downloaded the latest version, Tomcat 4.0.3 for Linux and untarred it.
> > When I launch it, I get an error that says : " Error unable to create
jar
> > cache in /tmp directory". I never encountered this error in any of the
> > previous versions of Tomcat I have downloaded and used, so this came as
a
> > surprise to me. Obviously, Tomcat cannot write a jar cache file to the
> /tmp
> > directory. I checked my file permissions to /tmp and everything seems OK
> > (i.e. readable, writeable globally.). So, anybody enocuntered the same
> error
> > before ? It doesn't affect any of the samples, or my previous servlets,
> but
> > it may cause problems down the road -- I'm not sure...
> >
> > Tomcat was run on a clean Turbolinux Workstation 7.0 install, on a Dell
> > system. Any help would be deeply appreciated.
> >
> > Regards,
> > pascal chong
> >
> >
> >
> >
> > --
> > 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]>

Unable to create jar cache

[Yu-Meng Chong]

Hi All !

I downloaded the latest version, Tomcat 4.0.3 for Linux and untarred it.
When I launch it, I get an error that says : " Error unable to create jar
cache in /tmp directory". I never encountered this error in any of the
previous versions of Tomcat I have downloaded and used, so this came as a
surprise to me. Obviously, Tomcat cannot write a jar cache file to the /tmp
directory. I checked my file permissions to /tmp and everything seems OK
(i.e. readable, writeable globally.). So, anybody enocuntered the same error
before ? It doesn't affect any of the samples, or my previous servlets, but
it may cause problems down the road -- I'm not sure...

Tomcat was run on a clean Turbolinux Workstation 7.0 install, on a Dell
system. Any help would be deeply appreciated.

Regards,
pascal chong




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

Re: Tomcat questions (long)

[Yu-Meng Chong]

Hi Josh !

I can help you out with the first part of your really long email (you
weren't kidding, were you ?). I, too, had a problem with all those
out.println()'s, and there *is* a better solution. You *can* try JSP code,
but it becomes difficult to maintain after a while -- especially if you're
trying to do some complex processing. My choice would be Velocity (see
http://jakarta.apache.org/velocity/index.html ) It's a template engine which
makes a very clean (IMHO, at least) division between interface and
implementation. How it works is : you create a normal HTML file, but put
placeholders in the code that will be populated by your servlet. Then you
write a servlet that does the processing and outputs the result as a string
which gets inserted into the placeholder. This is a VERY different approach
from the more "mainstream" method of JSPs and JavaBeans, which (again IMHO)
is too complicated and prone to error. A very good comparison can be found
here : http://jakarta.apache.org/velocity/ymtd/ymtd.html

As for the rest of your email ... well, I'll leave it to others more
knowledgeable and articulate than myself  to explain.

Regards,
Pascal Chong





- Original Message -
From: "JavaJosh" <[EMAIL PROTECTED]>
To: "Tomcat Users List" <[EMAIL PROTECTED]>
Sent: Monday, March 11, 2002 7:30 AM
Subject: Tomcat questions (long)


> Hi List,
>
> I am interested in building a data driven web app with Tomcat 4.0 and
> JSP/Servlets, but am new to the field, and have lots of questions. Some
> are general, some are specific to Tomcat. Please be patient, as I am
> new. Here it goes:
>
> System design: It seems that managing client state transitions is
> difficult. Unlike traditional GUIs, basically all forms are modal. How
> do you design around this? I have some ideas below that I would like
> some feedback on.
>
> Implementation: Looking at the servlet code that comes with Tomcat, it
> seems very ugly. All those out.println()'s give me the willies! JSP
> looks slightly better, since at least Java embedded in HTML is more
> readable than HTML embedded in Java! Even the latter can be minimized
> with taglibs, right?
>
> I have read some about FreeMarker and WebMacro. How exactly do these fit
> into the equation?
>
> Finally, regarding the design philosophy behind data-driven
> weblications: they are mostly parameterized finite state machines,
> right? There is a set of templates which, when requested, are filled
> with data, perhaps adjusted in some other, structural way, and served.
> Each user may have access to a different set of these templates (for
> example, limited by security). Each template may allow the user to
> transition to certain other templates. The process repeats during
> session lifetime.
>
> So it seems to me that a state diagram is in order, and each state
> corresponds to each HTML page the user sees. State transitions are
> basically hyperlinks. Because the number of states (pages) can grow
> large, it seems reasonable to expect them to be fairly uniform, and
> perhaps "inherit" in some way from a common parent template. Does this
> sound right? How can I achieve this?
>
> (this way of programming is much much different than the event driven
> concepts of, say, VB or Swing!)
>
> So a user logs in, gets a session identifier, and, unless all state
> transitions are allowed (the degenerate case), the server must also
> track client state, yes? Where can I find out more about how Tomcat
> assigns session identifiers and processes URLs to include them?
>
> If my understanding is correct, it seems that for most web apps 2
> servlets would be adequate: one for login and client instantiation
> (assigning session and initial state) and one for all other state
> transitions (call it servlet 2). After all, servlet 2 can get the
> session and then look up the client state from that, and then determine
> valid state transitions, and compare it to what the client requested to
> make sure everything is kosher! Logout would be a state transition on
> servlet 2, although there might be a polling process checking login time
> and auto-change user state when a certain amount of time has elapsed.
>
> I have a question about this and Tomcat: will servlet 2 be reused for
> all client requests? EG does Tomcat instance a copy of that servlet for
> every request, in the same way ServerSocket generates new Sockets? Or is
> the servlet required to behave linearly (that is, service request 1,
> then request 2, etc...)?
>
> One security concern would be spoofing the session data, probably after
> eavesdropping on the tcp stream/http content. This would only be of
> concern for the duration of the session, so it might pay to keep the
> sessions short. However, I thought of another thing to help prevent
> this: log the client IP address and correlate it with the session key.
> This won't help  in some circumstances, for example if the client is
> using NAT behind a firewall and the spoofer is behind there too (so they