eads makes the code
easier to maintain and the performance easier to scale when you move to
multi-processor systems. You need to judge that for yourself.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailt
ver
has closed the socket.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"mailto:[EMAIL PROTECTED]
My place on the web
s
not the only case of slight difference between up-event and
down-event handling between various platforms)
--
Michael SinzTechnology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web
es" all of that and provides a very nice, language
defined threading model with all of the tools needed to build complex
multi-threaded applications that run on any JVM of the same release
level. (JDK 1.2 is JDK 1.2 anywhere, for example)
--
Michael SinzTechnology and Engin
hat'd be bad.
Note that you can open a port and listen only on 127.0.0.1
(that is, not on all addresses but only on the specific address)
Since 127.0.0.1 is localhost it makes it relatively secure (unless
you let source-routed packets go to the 127.0.0.1 address, but
then you have othe
stem threading engine. Under Linux this is implemented
with the "clone()" system function and causes a process table entry for
each thread within the process. This is, my guess, what you are seeing
as "a new JVM starts" but is not really a new JVM but just a native L
.1.8 is very stable and is what I use on server applications.
(Swing and all of the user interface things are not used there)
I would say that by the time you upgrade there should be a stable final
release of the Java2 platform for Linux. Until then, JDK 1.1.7/1.1.8 are
your best bet.
--
Mich
1.1.7v3 you need to set
the NS_JAVA environment variable.
As of JDK 1.1.7v3, the JDK will also support noticing that you do not
have the DISPLAY environment set and thus will use the NS_JAVA setting.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"
nfo when running on the Xwin32 display?
(This is a simple console application) Please send the output of the
command as that will help identify the type of display environment X
thinks you have.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"
ther Java compilers
(such as Jikes) that produce different version IDs from what JavaC does.
(This is a tricky problem - again it should not be an issue as you should
you the same exact class file and not depend on recompiling)
--
Michael Sinz Technology and Engineering Director/Consultant
&
GUI) I don't know where that component would be useful.
Anyway, if you really need to create such an object you will need some form
of X environment, even if it is a simple frame buffer (such as Xvfb)
--
Michael Sinz Technology and Engineering Director/Con
e addressed this by doing things a bit differently
and thus should work in glibc 2.0 and glibc 2.1
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the
. (things like minor differences in the way certain routines work
which have major impacts on the JVM...)
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web ---> http://www.users.fast.net/~mic
hat broke it and we did not notice it until after
the release of RC1)
There is no need to try RC1 if you RC2 - they are otherwise the same.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the w
dows tool - the best thing to do is to redo the
operation under Linux with the tar tool that comes with Linux.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PR
ou
will still have glibc 2.0 along with glibc 2.1 and thus 117_v1a will work.
Also, 117_v3, while compatible with glibc 2.1 was linked against glibc 2.0
and thus if you have both on your system it will use glibc 2.0. This could
be your problem when using JNI.
--
Michael Sinz Technology a
pts will blindly use your settings
which means that if you get them wrong (or you change JVMs bot not all of the
settings) that you will cause things to fail.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECT
g package
naming.
Remember, Swing was an add-on in JDK 1.1.x Download the Swing 1.1.1 (which
is the current GA version) and add its swingall.jar to your classpath.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL
You most likely do not have that same environment running in the apache
server process or the servlet engine under apache (depending on how
you configured it and which servlet engine you used)
You should check that the correct environment (espcially locale and
timezone) settings exist in the web server
le. You need to make sure that
all of the files are readable/etc by the account you are running the Java
applications in.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROT
erence that you
never created an object for and then you try to do a method on the non-object
which is why you get the null pointer exception.
Try changing the ad=null to:
java.util.Vector ad = new java.util.Vector();
--
Michael Sinz Technology and Engineering Director/Consulta
the first assignment is valid)
In this case the a references an object that can be "cast" to a
object reference of class B and thus at run time it will work.
If you later do:
a=new A();
b=(B) a;
This would fail since a now contains an object that can no
iest to just use swingall.jar
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web ---> http://www.users.fast.net/~michael_sinz
-
TH environment variable
the java wrapper script will correctly add on the JDK system classes to
the class path.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web --
ch as winzip - would fail to
make the soft-links since Windows does not have such things)
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the w
native threads and,
some may claim, not 100% in green either...
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web ---> http://www.users.fast.net/~michael_sinz
what the bin/java
command is actually a soft-link to) will automatically find itself and
then set up the library load paths to include the JDK. If this does
not happen it may be due to the settings you have in JAVA_HOME or that
something got installed incorrectly on your system.
--
Michael
ve did you install? Also, did you double-check that
all files installed cleanly? (No errors, etc) and that the directory
you did this in supports soft-links?
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"
t the same as
creating the file) but that may not provide the final check for you.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web ---> http://www.users.fast.net/~michael_sinz
---
Now, you can scan the directory on your server via HTTP (or other
protocols your server may support) but you can only get back to
your own server this way (or at least you should only be able to get
back to your own server - that is part of the security)
--
Michael Sinz Technology and Engineerin
nd documentation and tutorials)
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
My place on the web ---> http://www.users.fast.net/~michael_sinz
---
tion, not just
>> on those that contend for the mutex.
>
>Is this really true?
>Why would an implementation that uses test-and-set at user-level
>and that would fall back on the kernel lock/unlock construct not be
>applicable in this case as well as in the signal-based im
be its own "process" in "ps" or "top"
This also means that the total number of threads in all running programs
can never be more than the process table limit in your machine (which is
defined at kernel compile time)
--
Michael Sinz Technology and Engineering Dire
or should it
>actually run correctly?
The debug version of the JDK is specifically to help debug programs.
This also means that you need to have any RMI code built for debugging.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" m
points to may change. In C/C++, if you want
both the point to be constant and what it points to be constant you
need a const pointer to a const char - aka "const char * const x"
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"
"/" vs "\" differences and the fact that Windows systems have
this really annoying drive letter thing.
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups" mailto:[EMAIL PROTECTED]
ry the Xvfb server and have that set for the JServ session with
DISPLAY set to point at it - mostlikely something like DISPLAY=:2.0 or
whatever display number you start you Xvfb running as.)
--
Michael Sinz Technology and Engineering Director/Consultant
"Starting Startups"
this publicly as others may find it useful.
note - this is all done in a *single* class that acts as a URL aware
class loader that includes basic authentication support - the goal was
to have this be as simple as possible.
--
Michael Sinz Technology and Engineering Director/Consultant
des may go away (will go away on all 16-bit compiled console
programs and any "family" programs and some 32-bit programs with a 16-bit
header)
You need to check what program you are trying to run and make sure it really
is a 32-bit console app. If it is not you will have far too ma
no source available - most likely due to some Sun-internal licensing of
technology from a 3rd party and thus they are not able to release the
source.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web
on't get any messages from the list itself.
Will do.
Like I said in the past on this list, the devil is in the details.
Getting the rough cut running was some work. However, you all should
know that the last 5% of a project takes 90+% of the effort. Polishing
and making stable are rat
ng java on
>linux. I have a glibc system, and also have libc-5 librarys installed.
>(Debian 2.1, 2.00.36)
>
>The tarball I downloaded is jdk_1.1.7-v1a-libc5-x86.tar.gz
Generally, if you have glibc you should use the glibc release of the
JDK.
Michael Sinz -- Director of Research & Dev
n RH6 or is it wise to
>get the source and compile it?
Please read the past postings to this list...
RedHat 6.0 is glibc 2.1 based and broke the 1.1.7 v1a release of the JDK.
We have made a 1.1.7 v3 release available that fixes this problem.
Michael Sinz -- Director of Research & Developmen
e is the diffs from the Sun source code - it is not
a diff from v1a to v3.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
On Mon, 31 May 1999 15:53:18 +0200, Per Widerlund wrote:
>Michael Sinz wrote:
>MS> Why? Java is platform independant and many (most) non-UNIX
>MS> platforms use other mechansims and, if they have a POSIX
>MS> layer, they emulate it only.
>
>If that is the case,
chanisms needed
to produce the results.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
---
3 (check your nearest blackdown
mirror site for the archives) which works in glibc 2.0 and 2.1
systems.
--
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the w
a minor benefit (that is, the fact that no decompiler
can decompile the result is a side effect and is not the reason I use it)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the we
On Sat, 22 May 1999 17:51:05 +1000, Steve Nguyen wrote:
>I tried to install JDK117_v3 into Slackware 3.6, Kernel 2.0.36 (libc5).
Note that JDK 1.1.7-v3 is a glibc binary. Older libc5 needs to continue
to use JDK 1.1.7-v1a (there is almost no difference between the two)
Michael S
t. Try just having classpath be ~/java
if you need it to be that.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
ge where as other VM's do it a
>little more smoothly.
The GC in our port is what is in Sun's code.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web --
ht now so I have very little time for the next while :-()
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
--
would be a *big* security hole (large enough to drive a truck
or two through without even looking)
Anyway, applets can do this if they are given the security clearance
which can happen with signed applets or applets run from local servers
with security turned off from the applet viewer.
1.7
(It is JDK 1.1.7 v3 currently)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
--
T
threads code has.
Also, at this time there is no native threads support for libc5 systems.
Only glibc based Linux systems.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place o
s. The fonts in the
JDK 1.2 pre release had some really ugly spacing characteristics.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
configuration
- Includes numerous bug fixes direct from Sun (1.1.7B)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http
quot; be. The code was
written in such a way that it could be anything and would change as
you changed the hardware (and in the Java world, as the JVM changed)
since there was nothing in the code to make it deterministic as to its
results.
Michael Sinz -- Director of Research & Development
you know exactly what the instructions are)
As such, a large time slice or a very fast bit of code would make your
little test have the 100 lines of output per thread happen all in one
timeslice. (Depending on the buffering in the output stream, this is
not hard to do in today's hardware)
Michae
ed to use in GLIBC 2.0 that is no longer available - a new release
should be showing up very soon now.)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web
just fine. (Well, as much as I tested it)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
-
tand-along applets"? Applets always had to be
run in the applet viewer or a browser.
Stand alone applications work just fine.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nex
o to change the .java_wrapper and put the directory
>> $JAVA_HOME/lib/org in the CLASSPATH of this script, but it still don't
>> work.
Again, you are pointing to the wrong place. You would need $JAVA_HOME/lib
since the "org" part is part of the class names themselves. A
ager
things) We are just getting a "roundtoit" for testing and releasing.
(BTW - we have had very few roundtoits. A few square ones but otherwise
not many at all - and most have been placed into the Java 2 pile which
seems to never have enough.)
Michael Sinz -- Director o
you please explain to us how we can run our java command line program
>> without worrying about silly things like X11 printer libraries? My
>> programmer is under the impression that we MUST install all this GUI crap
>> just to get out simple command line utility working. I on the
something wrong
>here?
This is not a bug in the JDK - you may wish to look at the getHostAddress()
method which will return a string of just the address.
See the java.net.InetAddress JavaDOCs for details.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailt
no longer exists in glibc 2.1...
In fact, Juergen has done the work already and I am just getting some time
to test it. (We have all been rather busy with other work...)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.
e not done so for the pre-release Java 2 environment.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
-
--troy
>
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
Michael Sinz -- Director of Research & Development, NextBus
e.
(Why have to make Jfoo, Nfoo, etc., when it really is a foo?)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
---
helpful and we would not have had the pre-release done as quickly
as it was without their support.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web --->
, what some would call, esoteric. They have to
do with race conditions in the threads and performance issues and the
like. (Threading on Linux is very different than Solaris and there are
some issues with the way some signals are handled and...) For most of us,
these types of issues are critical fo
"Compiled Code" is output due to the fact that you are running
a JIT of some sort. Run without it and you will get the line numbers.
It is also possible to build java byte code without line number information
in it. Usually this is done with "-O" in things like jikes and java
y
because of this. I normally have the wrapper script just adjust CLASSPATH
environment variable to make sure it works.
Now, in the JRE, the "-cp" does not need the core java classes...
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTEC
__
>Oliver Fels| e-mail:
>Neurotec Hochtechnologie GmbH | [EMAIL PROTECTED]
>Team Manager JAVA-/IT-Security |
>Friedrichshafen, Germany |
>-------
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROT
flags as many not-to-spec operations as possible. This would
address most of the "test everywhere" problems since you would
only need to test in the "strict" environment.
I wonder if such a tool is available or if such a tool would
be worth money :-)
Michael Sinz -- Di
eeded when this happens. Also, with -O it will end
up having to compile more than you thought due to some inlining.
Jikes does this a bit better (and it is sooo much faster that you
may not even notice if it were recompiling all files)
Michael Sinz -- Director of Research & Development, NextBu
the new fine-grain
security manager.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
ical scrollers
and it is handled by peers. This even works in JDK 1.0.2 and in applets.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web
On Sat, 27 Mar 1999 08:42:15 -0500, Michael Sinz wrote:
>>From: Bryce McKinlay
>>
>>Hi, I have a weird problem with JDK 1.2 - 'ps' and 'top' report massive amounts of
>memory being used by the jdk whenever I run *anything* in java:
>>
>>[bry
e that is running) is the byte code that
is running (and any JIT'ed native code if you happen to have a JIT too)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fas
is generally will not affect most people as far as Java is concerned
unless you are trying to pass the JCK tests, some of which test the
multicast socket support in Java (and thus found some bugs in Linux)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROT
ou sent it in since so many people were jumping
on the JDK 1.2 sources the moment Sun did an actual release.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www
ergeMem site has some examples as to JDK memory usage:
http://www.complang.tuwien.ac.at/ulrich/mergemem/
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web
are: The classes.zip
class files. These are loaded by the JVM into each JVM. This is not
shared automatically. However, there is hack to the VM system called
MergeMem that addresses this type of shared data too. I, personally,
do not use it but I have seen it work. The web site is:
htt
nd the XFree86 server compiled
for Win32. Both seem to run just fine.
The other problems will need to be looked at. The pre-release was
really a pre-beta - There are bound to be many issues that need to be
cleaned up (including install/etc)
Michael Sinz -- Director of Research & Develop
s need to
thing "both ways")
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
---
operations while the last one is still going, etc.
I would highly recommend it.
Check http://rufus.w3.org/linux/RPM/lftp.html for an RPM...
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the w
is label show up on the window.
Where is the addNotify()? (Just a guess... But that is what sets up
the peer...)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My pl
of code work?
I believe you need to make a simple class loader to add your own
path elements at run time from within Java. It is not a difficult
thing to do.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My
l
is called "srvany.exe" and it takes as options the thing to run.
Make that "java.exe classname options ..." with correct paths and
settings for the starting directory and it will work. I have run
some of our servers (for NextBus) on NT that way about a year ago
to show it coul
later.
The JDK 1.1.7v1 port will work just fine. You should use the libc5
version since RedHat 4.x is libc5 based. (RedHat 5.x is glibc - aka libc6)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place o
ease is installed you would need to remember
to clean up any cruft otherwise other confusion will occure.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.
to run.
The status of class files is based on your UMASK. I normally run 007
for company work and 002 for "fun" stuff.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
end the file to me from the web interface.
Normally not. A signed applet that the user lets the browser drop
the security for can do so but only then.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place o
On Fri, 19 Feb 1999 17:29:55 -0500, [EMAIL PROTECTED] wrote:
>No, I'm seeing the same problem with both green and native threads...
Hmmm... I run Java from shell scripts all the time.
Which shell are you using?
Michael Sinz -- Director of Research & Development, NextBus Inc.
m
elatively well but there are known limitations and
some other problems with it. You may have bumped into one or two of these.
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] - http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
bsolutely
required (as in, when doing admin work). This is a good habit to be
in since it tends to reduce your chances of really getting messed up
or having some unwanted access into your system. It also helps
identify things like access rights issues in your directories and files
since non-ro
as Java 1.1 built in and they have a JIT that is a bit better than TYA
in their JVM. (Even the Linux Netscape does on x86 - just like on
Windows x86 systems.)
Sun did much of the original activator work and it is not part of the
JDK 1.1 source license we had.
Michael Sinz -- Director of Resear
e problem is not that simple - there is the use of a few symbols
that are not longer available. (But then the whole ld wrapper stuff never
was something that I understood as to why Sun changed it.)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED
1 - 100 of 217 matches
Mail list logo
