Re: jdk1.2 testing
Trevor Harmon writes: > Gabe Hamilton wrote: > > > If you need people to test jdk 1.2, I would be happy to help out. > > Me too... It would be great if the Blackdown team could put out a > pre-release of 1.2, since their web page claims that the port is at > least "running". Anyone know if this is a possibility? We're not legally allowed to put out pre-release versions if they haven't passed through JCK. We're in the process of trying to chase down some issues with running the JCK. > The page also says, "Before we can release, we have to make sure that it > passes the tests in the Java Compatibility Kit." Is this a requirement > of Sun's source code licensing, or is it just something that was decided > on by the Blackdown team? It's the dreaded sections 2.4 and 2.5 as amending 2.3 of our license that say we HAVE to pass JCK before we can release or even pre-release. Steve
Java-VisiBroker CORBA on Linux with Blackdown Java 1.16/1.17 Howto
Hi,
after some mails about VisiBroker, here are some hints how to run
VisiBroker on Linux:
1) Everything from VisiBroker runs fine on Linux (pure Java), but not
the osagent process
which is nativ binary (and not supported on Linux)
2) if you don't use the name server and don't run the gatekeeper, you
don't need the osagent
and you can use a standalone Linux box.
You need then the URL naming scheme, - the server emits his IOR string,
the client grabs
and narrows it to use it.
3) if you want to start the gatekeeper (or naming service) on Linux, you
need to have an osagent
running somewhere on the network (on a NT, AIX, Sun or another supported
platfom)
Specify on your Linuxbox where to find the osagent:
a) For example, that's how i start the name server (similar i start my
servers)
java -DORBagentAddr=192.168.200.8 -DORBagentPort=14000
-DORBservices=CosNaming -DSVCnameroot=CorbaJava -DJDKrenameBug
com.visigenic.vbroker.services.CosNaming.ExtFactory CorbaJava nameLog
b) Servlets (jserv module with apache):
b1) the servlets.default.initArgs can be spcified in the servlet itself
like:
Properties props = System.getProperties();
props.put( "ORBservices", "CosNaming" );
props.put( "SVCnameroot", "CorbaJava" );
props.put( "ORBagentAddr", "192.168.200.8" );
props.put( "ORBagentPort", "14000" );
System.setProperties( props );
b2) or generally in (for example):
cp /usr/local/jserv/servlet.properties
/usr/local/apache/etc/servlet.properties:
and add a line:
servlets.default.initArgs=DefaultTemplDir=/usr/local/apache/share/templates/,ORBagentAddr=192.168.200.8,ORBagentPort=14000,ORBservices=CosNaming,SVCnameroot=CorbaJava
You access it then with:
Properties props = System.getProperties();
if (conf.getInitParameter("ORBagentAddr") != null)
props.put( "ORBagentAddr",
conf.getInitParameter("ORBagentAddr"));
System.setProperties( props );
c) Applets:
Without the netscape plugin in the tag:
http://192.168.200.8:9090/gatekeeper.txt>
When using the activator Java plugin for netscape communicator, use the
tag and for example:
http://192.168.200.8:9090"
org.omg.CORBA.ORBClass="com.visigenic.vbroker.orb.ORB"
ORBServices="CosNaming"
SVCnameroot="CorbaJava"
ORBagentAddr="192.168.200.8"
LOCdebug="true"
Hopefully that helps others
Marcel
Thanks to
Richard Jones <[EMAIL PROTECTED]>
Armen Yampolsky <[EMAIL PROTECTED]>
and a third person (Sorry i can't find your email any more)
which gave me this answers
--
Marcel Ruff
[EMAIL PROTECTED]
Java2
Hi folks -- Any words on when the Java 2 (aka 1.2) port is available? I would love to get a port of Jini out on Linux early (and often), but we are completely 1.2 based. Any words you could give would be a help. Thanks, Jim Waldo
Re: Java-VisiBroker CORBA on Linux with Blackdown Java 1.16/1.17 Howto
Marcel,
Thanks for your great summary! BTW, I just wanted to let you know, we've switched to
omniORB2/Sun IDL (JDK1.2) instead of Visibroker. I felt that visi was just way too
expensive, and being a believer in the benefits of Open Source software (and reading
all the great posts re omniORB), I decided to give it a try. The Sun java libs are
just enough for the client, definitely not good enough for a server, IMO, but we're
running a C++ appserver and so this fits beautifully. The two work together quite
well, and I only have good things to say about omniORB, so if you have similar
requirements, you may want to look into it. I guess this is sounding a bit like a
pitch ;)
Naturally, all work splendidly on linux (with the exception of the absence of
idltojava compiler, which the Blackdown team should have up Real Soon Now, right?).
Cheers,
-A.
Marcel Ruff wrote:
> Hi,
> after some mails about VisiBroker, here are some hints how to run
> VisiBroker on Linux:
>
> 1) Everything from VisiBroker runs fine on Linux (pure Java), but not
> the osagent process
> which is nativ binary (and not supported on Linux)
>
> 2) if you don't use the name server and don't run the gatekeeper, you
> don't need the osagent
> and you can use a standalone Linux box.
> You need then the URL naming scheme, - the server emits his IOR string,
> the client grabs
> and narrows it to use it.
>
> 3) if you want to start the gatekeeper (or naming service) on Linux, you
> need to have an osagent
> running somewhere on the network (on a NT, AIX, Sun or another supported
> platfom)
> Specify on your Linuxbox where to find the osagent:
>
> a) For example, that's how i start the name server (similar i start my
> servers)
> java -DORBagentAddr=192.168.200.8 -DORBagentPort=14000
> -DORBservices=CosNaming -DSVCnameroot=CorbaJava -DJDKrenameBug
> com.visigenic.vbroker.services.CosNaming.ExtFactory CorbaJava nameLog
>
> b) Servlets (jserv module with apache):
>
> b1) the servlets.default.initArgs can be spcified in the servlet itself
> like:
>
> Properties props = System.getProperties();
> props.put( "ORBservices", "CosNaming" );
> props.put( "SVCnameroot", "CorbaJava" );
> props.put( "ORBagentAddr", "192.168.200.8" );
> props.put( "ORBagentPort", "14000" );
> System.setProperties( props );
>
> b2) or generally in (for example):
> cp /usr/local/jserv/servlet.properties
> /usr/local/apache/etc/servlet.properties:
> and add a line:
>
>servlets.default.initArgs=DefaultTemplDir=/usr/local/apache/share/templates/,ORBagentAddr=192.168.200.8,ORBagentPort=14000,ORBservices=CosNaming,SVCnameroot=CorbaJava
>
> You access it then with:
> Properties props = System.getProperties();
> if (conf.getInitParameter("ORBagentAddr") != null)
>props.put( "ORBagentAddr",
> conf.getInitParameter("ORBagentAddr"));
>
> System.setProperties( props );
>
> c) Applets:
> Without the netscape plugin in the tag:
>
>
>
>
> value=http://192.168.200.8:9090/gatekeeper.txt>
>
> When using the activator Java plugin for netscape communicator, use the
> tag and for example:
>type="application/x-java-applet;version=1.1"
> width="844"
> height="468"
> align="middle"
> CODE="org.freelancer.fancy.class"
> CODEBASE="http://192.168.200.8:9090"
> org.omg.CORBA.ORBClass="com.visigenic.vbroker.orb.ORB"
> ORBServices="CosNaming"
> SVCnameroot="CorbaJava"
> ORBagentAddr="192.168.200.8"
> LOCdebug="true"
>
>
> Hopefully that helps others
>
> Marcel
>
> Thanks to
> Richard Jones <[EMAIL PROTECTED]>
> Armen Yampolsky <[EMAIL PROTECTED]>
> and a third person (Sorry i can't find your email any more)
> which gave me this answers
>
> --
> Marcel Ruff
> [EMAIL PROTECTED]
--
Armen Yampolsky
Axiom Software Labs
New York
(possible bug) "Invalid argument" in native code
Hi! The following problem exists in both JDK 1.1.6v5, and 1.1.7v1a. I've got a glibc systems, and using the glibc versions of the JDK. My kernel is 2.0.34. I'm pretty sure this is Linux specific, but I haven't been able to verify that by trying it on another OS. But here goes. Under certain circumstances which I will explain below, ServerSocket.accept() results in a SocketException, with the message "Invalid argument", thrown in PlainSocketImpl.java, line 379 (line 387 in JDK 1.1.6v5). That method calls the native method socketAccept(SockImpl s) (I checked the JDK enclosed Java source code). Now, what exacly is going wrong in the underlaying implementaion, when it reports "invalid argument"? It sounds to me like the native implementation passes some weird argument to listen() or something similar. That, I think, would indicate a JDK bug. Here's the scenario in which this happens: I have a server that, among other things, goes into a loop, accepting clients with ServerSocket.accept() (what else?) and lets another thread do its thing with the returned Socket. I also have a client that, among other things, connects to the server, with a "... new Socket(hostname, 800)" call. All is fine, and it works. I now kill the server, and the client will detect the disconnected state of the Socket by the -1 returned by InputStrea.read(), and will reconnect as follows, plus some waiting and exception handling of course: sock.close(); // Close client's end of the socket sock = new Socket(hostname, 800); // IOException: Connection refused Here's where the problem comes in. The client's connection attempt will fail. On the client side, an IOException with the message "Connection refused" is thrown. On the Server side, ServerSocket.accept() has now resulted in the mysterious SocketException mentioned above. Onte last note: The client attempt to connect to the server every 10 seconds. If I make the server repeatedly try to call .accept() again, disregarding the exceptions, a handfull of those calls will result in the same SocketException. Then it will stop throwing them and start blocking again - until the clients attempts another reconnect. It all works perfectly if I shut down the client, start it up again and reconnect. No problem. It's only when the client is trying to reconnect after an unexpected disconnect that the problem occurrs. I am unable to figure out what I am doing wrong. Frankly, I don't have a clue. 1) Could this be a JDK bug? 2) Does anyone know how I might fix it? (if it's not Linux related, I guess that should go off the mailinglist) Thanks! / Peter Schuller E-Mail: [EMAIL PROTECTED] Web: http://hem.passagen.se/petersch
STOP asking about Java 1.2 / 2
I know we're all holding our breath while waiting for it, I know we're all so enthusiastic about it. I know that we read everywhere about Java 2 and we don't want to use winbug to try it. I'm waiting for it like all of you. But *please* STOP ASKING about Java 1.2/2. First of all it's a FAQ, Blackdown have already answered: in a month. Yes, in a month, but what was the first day of this month? This a good chance to learn a bit of HTTP, I say: instead of asking, find a way to know it by yourself. You don't want to learn HTTP, wait a month from NOW or check the page every day. The Blackdown people is working on it, they will not give you a pre-release, because if they could they would have done it already. Are you so eager? Start downloading the 1.2 documentation and study it. Let the Blackdown Organization work on the Linux port in peace! Damn, they are doing it for well, they are doing it for free...what else do you want? A cup of Java? Oh, yes, now I wait until the JDK 1.2 Linux port is released and the same people begin complaining that is buggy (later we'll find out it's due to Javasoft) and ask why it was released before it was carefully tested. I'm very annoyed by the lack of respect for the Blackdown's work. --- Andrea "Kontorotsui" Controzzi - MALE Student of Computer Science at University of Pisa - Italy - E-mail: [EMAIL PROTECTED] My home page: http://www.cli.di.unipi.it/~controzz/intro.html Founder and Admiral of Hoshi no Senshi (italian Leiji Matsumoto's fan group). Creator of It.Arti.Cartoni (italian anime newsgroup) and proud member of... +-+ |. * . | | .__ . . | |oq |po _ _| | / #==>>>==#,-' (_)\ | | | ,-|~\\ ///_ ,() ,_} | | | |/|~]]] /// ,-~' .,~ / \| . | | |\_|_|_\_\~~~~' \ (/|. | | ./~ \___/ [m] \ \__// | | _bo..__ // `-,.~~ | | _-~ 0o.__( . | | \ o . | | . (_)00 | |. \~~~*,,,* ~00 | |~0 . | | ~~~---~~ | | .*| +-+
Re: help can't find HelloWorld class message
You need to amke sure . is in your classpath. Then you need to type the name of the program without the .java extension. Here is what works on my system. % setenv CLASSPATH . % java HelloWorld Mo DeJong dejong at cs.umn.edu On Fri, 8 Jan 1999, Hoang C. Truong wrote: > Hello, > When I compile my java source using this command: > javac HelloWorld.java > It produced HelloWorld.class. > When I try to run it, it gave me error message like this: > java HelloWorld > can't find HelloWorld class > Does anyone know what is going on? Thanks > > >
Re: jdk 1.2 was released too early
On Fri, 8 Jan 1999, Kirk Hutchinson wrote: > For those of you who are not programming on multiple platforms and > have not begun using the much-balleyhooed Java 2, let me clue you in on > a few things. I know exactly what you mean. I am working on a project that combines Tcl and Jacl and I can tell you that the JDK1.2 has a number of really nasty bugs. Heck, they even broke charcter echo when typing into a DOS shell! > Java 2 was released WAY too early, and consequently suffers from > many bugs and problems. The following parts are things my team > has encountered where intelligence has changed for the worse, or > is just plain broken now: > > combo boxes > table models > character consuming > JNI - still sucks - especially on Solaris > Metal look and feel - color and image problems, wicked spacing Tell me about it, Runtime.exec needs to be fixed in a big way. If you want to see my suggestion for a fix check out bug # 4156278 at the developer.javasoft.com site. > System.exec problems still > > There are a number of others crucial to our project that have > been on hold due to Javasoft's negligence. > > Plus, there are a number of issues which have not been resolved that > should have been over a year ago. > > There are a few plusses (RMI is faster, overall view speed has improved), but > the negatives are definitely more than the positives. > I can't wait for 1.2.1! Well, the problem with that is that they did not reall "open" up the source. The source is a little more open but Sun still retains all control of the spec and you can not use the source code without paying Sun. later mo dejong at cs.umn.edu > Personally, I think they opened the source because they realized they > were falling so too behind the performance and ability they were > supposed to have by now. The more open licensing will hopefully allow > others to fix areas of great need they just haven't fixed yet. > > Stick with 1.1.7 and you'll be better off until 1.2.1 (or 2.0.1 - if > they can ever get their names straight). > > Java will be great if they don't keep screwing it up! > > kirk > -- > Kirk Hutchinson, [EMAIL PROTECTED] > Electrical & Software Engineer, Cabletron Systems > What good is unused science? >
Re: jdk1.2 testing
Is there any way we can send you some code so that you can test it out in the 1.2 release? I have some native threads + JNI code that works under Solaris and Windows JDK but breaks with the 1.1.7 Native threads release. I know for a fact that the code is correct but there is some kind of problem with the 1.1.7 version of the JNI_CreateJavaVM function. I know you guys do not want to get into a situation where you are testing a lot of code, but could you consider it for code that is know to not work in the 1.1.7 release? I really have no way to debug the JNI_CreateJavaVM function without the source code. thanks mo dejong dejong at cs.umn.edu On Fri, 8 Jan 1999, Steve Byrne wrote: > Trevor Harmon writes: > > Gabe Hamilton wrote: > > > > > If you need people to test jdk 1.2, I would be happy to help out. > > > > Me too... It would be great if the Blackdown team could put out a > > pre-release of 1.2, since their web page claims that the port is at > > least "running". Anyone know if this is a possibility? > > We're not legally allowed to put out pre-release versions if they haven't > passed through JCK. We're in the process of trying to chase down some issues > with running the JCK. > > > The page also says, "Before we can release, we have to make sure that it > > passes the tests in the Java Compatibility Kit." Is this a requirement > > of Sun's source code licensing, or is it just something that was decided > > on by the Blackdown team? > > It's the dreaded sections 2.4 and 2.5 as amending 2.3 of our license that say > we HAVE to pass JCK before we can release or even pre-release. > > Steve >
Re: Java-VisiBroker CORBA on Linux with Blackdown Java 1.16/1.17 Howto
Sorry if this is off-topic, but you might also look at ORBacus. I think it was based on OmniORB originally, but is being commercially developed. Their prices are reasonable, and they license it for free for non-commecial uses. The Java ORB works fine on Java 1.1.7 on linux. We've been using it with great success for about 4 months now. http://www.ooc.com/ Kelly -- Kelly A. Campbell Applications Programmer/Analyst <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> Kansas State University http://www.telecom.ksu.edu/~camk/ Department of Telecommunications 109 East Stadium, Manhattan KS 66506http://www.telecom.ksu.edu/ Voice: (785) 532-7067 Fax: (785) 532-7114 On Sat, Jan 09, 1999 at 10:50:29AM -0500, Armen Yampolsky wrote: > Marcel, > > Thanks for your great summary! BTW, I just wanted to let you know, we've switched to >omniORB2/Sun IDL (JDK1.2) instead of Visibroker. I felt that visi was just way too > expensive, and being a believer in the benefits of Open Source software (and reading >all the great posts re omniORB), I decided to give it a try. The Sun java libs are > just enough for the client, definitely not good enough for a server, IMO, but we're >running a C++ appserver and so this fits beautifully. The two work together quite > well, and I only have good things to say about omniORB, so if you have similar >requirements, you may want to look into it. I guess this is sounding a bit like a >pitch ;) > > Naturally, all work splendidly on linux (with the exception of the absence of >idltojava compiler, which the Blackdown team should have up Real Soon Now, right?). > > Cheers, > -A.
Re: jdk1.2 testing
I recently tried to create a C app that created a JVM, started a simple Java class, and passed messages from C (or C++) to Java. After running successfully on NT, I noticed that it core dumped when it was created from within another C executable on Solaris. When I used Purify on it, EVERY call to the JNI API was flagged with problems, and the core dump was happening from a JNI call (unfortunately, I can't remember exactly which one...). More than likely, the Linux JNI will suffer the same fate as the Solaris JNI. Pity. kirk Moses DeJong wrote: > > Is there any way we can send you some code so that you can test it out > in the 1.2 release? I have some native threads + JNI code that works under > Solaris and Windows JDK but breaks with the 1.1.7 Native threads release. > I know for a fact that the code is correct but there is some kind of > problem with the 1.1.7 version of the JNI_CreateJavaVM function. I know > you guys do not want to get into a situation where you are testing a lot > of code, but could you consider it for code that is know to not work in > the 1.1.7 release? I really have no way to debug the JNI_CreateJavaVM > function without the source code. > > thanks > mo dejong > dejong at cs.umn.edu > > On Fri, 8 Jan 1999, Steve Byrne wrote: > > > Trevor Harmon writes: > > > Gabe Hamilton wrote: > > > > > > > If you need people to test jdk 1.2, I would be happy to help out. > > > > > > Me too... It would be great if the Blackdown team could put out a > > > pre-release of 1.2, since their web page claims that the port is at > > > least "running". Anyone know if this is a possibility? > > > > We're not legally allowed to put out pre-release versions if they haven't > > passed through JCK. We're in the process of trying to chase down some issues > > with running the JCK. > > > > > The page also says, "Before we can release, we have to make sure that it > > > passes the tests in the Java Compatibility Kit." Is this a requirement > > > of Sun's source code licensing, or is it just something that was decided > > > on by the Blackdown team? > > > > It's the dreaded sections 2.4 and 2.5 as amending 2.3 of our license that say > > we HAVE to pass JCK before we can release or even pre-release. > > > > Steve > > -- Kirk Hutchinson, [EMAIL PROTECTED] Electrical & Software Engineer, Cabletron Systems What good is unused science?
Re: STOP asking about Java 1.2 / 2
On Sat, 9 Jan 1999, Kontorotsui wrote: > > But *please* STOP ASKING about Java 1.2/2. Don't be silly. Nothing you say will stop new subscribers asking the question most on their minds. > > First of all it's a FAQ, Blackdown have already answered: in a month. > Yes, in a month, but what was the first day of this month? When last I looked this statement was undated. Someone here menioned in the past few days that it's still undated. afaicr it's a month or so old now. > This a good chance to learn a bit of HTTP, I say: instead of asking, find a > way to know it by yourself. You don't want to learn HTTP, wait a month from > NOW or check the page every day. Now you're being rude, crude and unattractive. Well, maybe not crude, but certainly patronising. This one of the few ways you can make me feel inclined to feed you a bunch of fives. > > The Blackdown people is working on it, they will not give you a pre-release, > because if they could they would have done it already. > > Are you so eager? Start downloading the 1.2 documentation and study it. Let > the Blackdown Organization work on the Linux port in peace! > Damn, they are doing it for well, they are doing it for free...what else do > you want? A cup of Java? > > Oh, yes, now I wait until the JDK 1.2 Linux port is released and the same > people begin complaining that is buggy (later we'll find out it's due to > Javasoft) and ask why it was released before it was carefully tested. > > I'm very annoyed by the lack of respect for the Blackdown's work. Any you should have regard for the feelings of others. I find the tone of your email offensive even though it wasn't directed to me. It is absolutely normal behaviour for people to ask for information. You should not assume everyone is up to your skills, or that everyone knows of the existance (or location) of the FAQ. Neither should you assume that it's convenient for everyone to retrieve the FAQ: I for one read most info offline and am not at my computer for most of the time it's online. -- Cheers John Summerfield http://os2.ami.com.au/os2/ for OS/2 support. Configuration, networking, combined IBM ftpsites index.
suggestion for list (was: Re: STOP asking about Java 1.2 / 2)
The recent on-list exchange between John Summerfield and Kontorotsui has
brought a background thought of mine to the foreground, so I'd like to share
it with the list and see what the general opinion is.
I suggest that the current copy of the FAQ be included with the succesful
subscription message. I know this is possible with majordomo because other
lists do it. I believe this will reduce the volume of these questions because
it appears that most of the people posting this question have just joined
the list. (probably in order to ask that question.)
I also suggest that all messages from the list have a standard footer
appended; something like what Redhat does on their lists. I'd suggest:
(Java && Linux) - [EMAIL PROTECTED] - !(Java || Linux)
To unsubscribe: issue the following command from a shell prompt
mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
or maybe that first line should be:
if (Subject == (Java && Linux)) { To = [EMAIL PROTECTED] };
Comments? Cheers all! -=Chris
<*> cabbey at rconnect dot com http://homepage.rconnect.com/cabbey/ <*>
Get it up, keep it up... LINUX: Viagra for the PC. ;) PGP Info: pgp.html
-BEGIN GEEK CODE BLOCK-
Version: 3.12 http://www.geekcode.com
GCS$/IT/PA$ d(-) s++:+ a-- C+++$ UL UA++$ P++ L++ E- W++ N+ o? K? !P
w---(+)$ O- M-- V-- Y+ PGP+ t--- 5++ X+ R tv b+ DI+++ D G e++ h(+) r@ y?
--END GEEK CODE BLOCK--
