glibc2/jdk1.1.5v7
OK... obviously I've not got them working together, so before I launch into full diagnostics: I notice on the website that glibc-2.0.7-7 is the redhat package recommended. Now, not being an rpm type, I only have 2.0.6 (this is the latest one I can *find*) which I compiled myself... So my question is, _could_ this be the reason for the segfaulting? Brief description: sounds a bit like what our friend Travis Shirk was going through; i.e., "java Class" segfaults, as does "javac". I've done a fair amount of tweaking and diagnostic work which I can post if anyone cares... I'd appreciate any hints/clues/suggestions andy - Andy Stubbs +44 151 794 4045 [EMAIL PROTECTED]
Re: glibc2/jdk1.1.5v7
Andy Stubbs wrote: > > OK... obviously I've not got them working together, so before I launch > into full diagnostics: I notice on the website that glibc-2.0.7-7 is the > redhat package recommended. Now, not being an rpm type, I only have 2.0.6 > (this is the latest one I can *find*) which I compiled myself... So my > question is, _could_ this be the reason for the segfaulting? > > Brief description: sounds a bit like what our friend Travis Shirk was > going through; i.e., "java Class" segfaults, as does "javac". I've done a > fair amount of tweaking and diagnostic work which I can post if anyone > cares... > > I'd appreciate any hints/clues/suggestions > > andy > > - > Andy Stubbs +44 151 794 4045 > [EMAIL PROTECTED] I installed jdk1.1.5v7 a few days ago, and upon testing got segfaults running javac. Also noticed the upgrade recommendation, (and having RH5.0, using an rpm) upgraded from whatever to glibc-2.0.7-7, _no_ problems since then... HTH. Bob L. -- Robert Lynch-Berkeley CA [EMAIL PROTECTED] http://www.best.com/~rmlynch/
Dynamic loading and libgcc need?
First, apologies if these topics have been beaten to death. I have had a hard time staying subscribed to this list as it seemed to bounce around from server to server, and can't seem to find any recent archives. Is it possible to specify a path for System.loadLibrary()? I have not been able to get LD_LIBRARY_PATH to have any effect. I hacked a version of java.lang.Runtime that dumps the result of the initializeLinkerInternal call to stderr and it appears to be consistently empty regardless of the value of LD_LIBRARY_PATH. I have also been having problems using javah when .zip files are in the classpath, entailing complaints about inability to resolve various symbols such as __muldi3. I suspect this is a result of not having -lgcc in the build line when building the shared libraries (notably libjava and libzip). I have not been able to test this theory, not having access to the original JDK source. ldd -r on libjava.so complains about missing libgcc symbols on both my Debian 1.3.1 (libc5 5.4.33) and pre-2.0 (libc6 2.0.6). I appreciate any insight or advice anyone can provide. -- Robert Coie Implementor, Intrigue Ltd.
Re: glibc2/jdk1.1.5v7
I belive glibc2.0.7-7 is in the ftp://ftp.redhat.com/pub/contirb/hurricane/i386 I would check but I am useing MacOS (not mine) and it is so messed up that I can no connect to some servers on the internet. > >OK... obviously I've not got them working together, so before I launch >into full diagnostics: I notice on the website that glibc-2.0.7-7 is the >redhat package recommended. Now, not being an rpm type, I only have 2.0.6 >(this is the latest one I can *find*) which I compiled myself... So my >question is, _could_ this be the reason for the segfaulting? > >Brief description: sounds a bit like what our friend Travis Shirk was >going through; i.e., "java Class" segfaults, as does "javac". I've done a >fair amount of tweaking and diagnostic work which I can post if anyone >cares... > >I'd appreciate any hints/clues/suggestions > >andy > >- >Andy Stubbs +44 151 794 4045 >[EMAIL PROTECTED] > > > > > > __ Get Your Private, Free Email at http://www.hotmail.com
Re: Linux/Alpha port of JDK
Once upon a time, James Cribb wrote > Chris Adams wrote: > > The biggest problem is that I can't run anything that uses the > > InstallShield Java Edition. > > After I installed the Intel Linux JDK 1.1.5 (v7), the Install Shield stuff in > Together/J (http://www.oi.com) seemed to activate a bug that killed my window > manager (ctwm) and hence shut down my X windows session. To work around this, I > started a separate X server without a window manager (using the command "X :1") > and ran the Together/J install program on that X server (by setting DISPLAY=:1 > in my environment). > > Maybe this will help in your situation. I have a different problem. X or the window manager doesn't crash. When I run an InstallShield installation script, I get: * Testing /usr/local/jdk-1.1.5/bin/java... Testing /usr/local/jdk-1.1.5/bin/jre... 2 suitable java interpreter(s) detected 0) I want to specify a path to an interpreter. 1) Use /usr/local/jdk-1.1.5/bin/java 2) Use /usr/local/jdk-1.1.5/bin/jre 3) Terminate this installation. Select a choice [0-3]: 2 Extracting installation class InstallShield JavaTM Edition Extracting installation codedone SIGSEGV 11* segmentation violation Full thread dump: "AWT-Motif" (TID:0x29d3890, sys_thread_t:0x200024bddb0, state:MW) prio=5 java.lang.Thread.run(Thread.java) "AWT-Input" (TID:0x29d38d0, sys_thread_t:0x2000249bdb0, state:CW) prio=5 "AWT-EventQueue-0" (TID:0x29d3900, sys_thread_t:0x20ffdb0, state:CW) prio=5 java.lang.Object.wait(Object.java) java.awt.EventQueue.getNextEvent(EventQueue.java:119) java.awt.EventDispatchThread.run(EventDispatchThread.java:55) "Finalizer thread" (TID:0x29cdab0, sys_thread_t:0x20dddb0, state:CW) prio=1 "Async Garbage Collector" (TID:0x29cdb40, sys_thread_t:0x20bbdb0, state:R) prio=1 "Idle thread" (TID:0x29cd620, sys_thread_t:0x2099db0, state:R) prio=0 "Clock" (TID:0x29c6110, sys_thread_t:0x2077db0, state:CW) prio=12 "main" (TID:0x29c6160, sys_thread_t:0x120141e70, state:R) prio=5 *current thread* sun.awt.motif.MComponentPeer.setBounds(MComponentPeer.java:467) sun.awt.motif.MComponentPeer.reshape(MComponentPeer.java:512) sun.awt.motif.MComponentPeer.initialize(MComponentPeer.java:91) sun.awt.motif.MTextFieldPeer.initialize(MTextFieldPeer.java:67) sun.awt.motif.MComponentPeer.(MComponentPeer.java:126) sun.awt.motif.MTextFieldPeer.(MTextFieldPeer.java:71) sun.awt.motif.MToolkit.createTextField(MToolkit.java:70) java.awt.TextField.addNotify(TextField.java:146) java.awt.Container.addNotify(Container.java:1128) java.awt.Panel.addNotify(Panel.java:77) java.awt.Container.addNotify(Container.java:1128) java.awt.Panel.addNotify(Panel.java:77) java.awt.Container.addNotify(Container.java:1128) java.awt.Panel.addNotify(Panel.java:77) java.awt.Container.addNotify(Container.java:1128) java.awt.Panel.addNotify(Panel.java:77) java.awt.Container.addNotify(Container.java:1128) java.awt.Panel.addNotify(Panel.java:77) java.awt.Container.addNotify(Container.java:1128) java.awt.Window.addNotify(Window.java:108) Monitor Cache Dump: sun.awt.motif.MToolkit@r": owner "main" (0x120141e70, 1 entry) Waiting to enter: "AWT-Motif" (0x200024bddb0) java.awt.EventQueue@it@r": Waiting to be notified: "AWT-EventQueue-0" (0x20ffdb0) java.lang.Object@": owner "main" (0x120141e70, 6 entries) (0x20bbdb0): owner "Async Garbage Collector" (0x20bbdb0, 1 entry) Registered Monitor Dump: Verifier lock: Thread queue lock: Name and type hash table lock: String intern lock: JNI pinning lock: JNI global reference lock: BinClass lock: Class loading lock: Java stack lock: Code rewrite lock: Heap lock: Has finalization queue lock: Finalize me queue lock: Waiting to be notified: "Finalizer thread" (0x20dddb0) Monitor IO lock: Child death monitor: Event monitor: I/O monitor: owner "AWT-Input" (0x2000249bdb0, 1 entry) Alarm monitor: Waiting to be notified: "Clock" (0x2077db0) Monitor registry: owner "main" (0x120141e70, 1 entry) Thread Alarm Q: * And it exits. It doesn't matter if I use 'java' or 'jre' as the interpreter. I am trying to run Amber, a system monitoring program for Livingston/Lucent RABU PortMaster remote access servers and routers. It is available at ftp://ftp.livingston.com/pub/le/software/java/. They now also have a .jar file that does not use InstallShield (it can be run directly with 'jre -cp . amber.jar Amber') and it runs, so I think the problem is with something that InstallShield does. -- Chris Adams - [EMAIL PROTECTED] System Administrator - Renaissanc
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998, charles verge wrote: > I belive glibc2.0.7-7 is in the > ftp://ftp.redhat.com/pub/contirb/hurricane/i386 > > I would check but I am useing MacOS (not mine) and it is so messed up > that I can no connect to some servers on the internet. Actually, the glibc 2.0.7-7 of which I speak is an official update for RH 5.0. It can be found in ftp://ftp.redhat.com/pub/redhat/redhat-5.0/updates/i386/ . Until RH gets ContribNet (http://developer.redhat.com/contribnet/) setup anything in contrib is use at your own risk. - cls
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998, charles verge wrote: > I belive glibc2.0.7-7 is in the > ftp://ftp.redhat.com/pub/contirb/hurricane/i386 > > I would check but I am useing MacOS (not mine) and it is so messed up > that I can no connect to some servers on the internet. As it happens, it isn't actually there; it's in the updates directory for RH 5.0. BUT... I can't work out why jdk1.1.5 is linked against a version of glibc2 which doesn't appear to exist in anything other than .rpm format and isn't even at the gnu ftp site. This is something of a mystery to me. I guess if I want this thing to work I'm going to either have to find a rpm depackager thingy and try to botch the result into the system somewhere without breaking anything, or give up and get libc5.4.44... which defeats the whole point somewhat. is a .tgz available _anywhere_ (archie thinks not)? andy > > > > > >OK... obviously I've not got them working together, so before I launch > >into full diagnostics: I notice on the website that glibc-2.0.7-7 is > the > >redhat package recommended. Now, not being an rpm type, I only have > 2.0.6 > >(this is the latest one I can *find*) which I compiled myself... So my > >question is, _could_ this be the reason for the segfaulting?
Re: apache 1.2.4 + mod-jserv + SdK
> I was wondering if anyone out there is trying the same as I am and that > is to set up the apache webserver for linux with the servlet module and > development kit voor java servlets. I've been trying to do some things > with the new jserv release of java and get it recompiled into a new make > of apache 1.2.4 for linux, but unsuccesful sofar. If anyone knows how to > get it to work I woul;d greatly appreciate it. I'm running RedHat 5 now, > but I can reinstall into Slackware anytime , if that will work. I was able to get it to compile as a shared module under Apache 1.3b6 under both Debian 1.3.1 (libc5) w/JDK1.1.5v7 and Debian pre-2.0 (libc6) w/JDK1.1.5-v5g. There are instructions in the README.DSO file in the Apache 1.3b6 source distribution with examples of how to configure external modules. That worked well, except I had to change a good deal of function names from functionXXX to ap_functionXXX (I assume Apache underwent namespace policy changes recently). I noticed that there is a compatibility header with #defines from the old names to the new ones, so that may be another approach. Now if I can only figure out how the !#$!@#!!& classloader works in jserv, because it seems to be wreaking havoc with my attempts to use JPL (Larry Wall's hybrid Java-Perl language) in servlets. The shared libraries load correctly, but I get UnsatisfiedLinkErrors trying to call things in them. -- Robert Coie Implementor, Intrigue Ltd.
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998, Andy Stubbs wrote: > > OK... obviously I've not got them working together, so before I launch > into full diagnostics: I notice on the website that glibc-2.0.7-7 is the > redhat package recommended. Now, not being an rpm type, I only have 2.0.6 > (this is the latest one I can *find*) which I compiled myself... So my > question is, _could_ this be the reason for the segfaulting? > > Brief description: sounds a bit like what our friend Travis Shirk was > going through; i.e., "java Class" segfaults, as does "javac". I've done a > fair amount of tweaking and diagnostic work which I can post if anyone > cares... I'd like to see what you you've done so far, because I feel like I've tried everything. I now have libc-5.4.44 and glibc-2.0.7-7 and still cannot get jdk1.1.5_v7 working. Core dump when running javac and java with no arguments. > > I'd appreciate any hints/clues/suggestions > > andy > > - > Andy Stubbs +44 151 794 4045 > [EMAIL PROTECTED] > > > > > Travis -- Travis Shirk [EMAIL PROTECTED]
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998, Christopher Seawood wrote: [ [Actually, the glibc 2.0.7-7 of which I speak is an official update for RH [5.0. It can be found in [ftp://ftp.redhat.com/pub/redhat/redhat-5.0/updates/i386/ . Until RH gets [ContribNet (http://developer.redhat.com/contribnet/) setup anything in [contrib is use at your own risk. Just in case anyone hasn't said it yet, add my name to the list of the eternally grateful for the glibc jdk, Chris,Steve, Karl et al. =joe - Joe Murray | Systems | O'Dell Engineering Ltd. | Voice: 519 740 8620 | Fax: 519 740 9483 | [EMAIL PROTECTED] | -
Re: Dynamic loading and libgcc need?
On Tue, May 12, 1998 at 10:35:14AM -0700, Robert Coie wrote:
> Is it possible to specify a path for System.loadLibrary()? I have not
> been able to get LD_LIBRARY_PATH to have any effect. I hacked a
> version of java.lang.Runtime that dumps the result of the
> initializeLinkerInternal call to stderr and it appears to be
> consistently empty regardless of the value of LD_LIBRARY_PATH.
System.load("/path/to/library.so");
Regarding your LD_LIBRARY_PATH problems, are you sure your environment
has it? If (for instance) you run your program under some shell spawned
by your editor you need to make sure the shell reads the proper init
file for _non-interactive_ use. For example zsh only reads .zshenv when
used from an editor (that's where I keep CLASSPATH, LD_LIBRARY_PATH,
etc..).
> I have also been having problems using javah when .zip files are in
> the classpath, entailing complaints about inability to resolve various
> symbols such as __muldi3. I suspect this is a result of not having
> -lgcc in the build line when building the shared libraries (notably
> libjava and libzip). I have not been able to test this theory, not
> having access to the original JDK source. ldd -r on libjava.so
> complains about missing libgcc symbols on both my Debian 1.3.1 (libc5
> 5.4.33) and pre-2.0 (libc6 2.0.6).
Never had this problem, despite using javah regularly (Debian-2.0). Some
library version mismatch probably.
--
Louis-David Mitterrand
http://www.aparima.com
[EMAIL PROTECTED]
Re: glibc2/jdk1.1.5v7
PLEASE DELETE THE INCOGNITO.NET DOMAIN FROM THIS LIST. PLEAE ADVISE HOW IT BECAME LISTED. -Original Message-From: charles verge <[EMAIL PROTECTED]>To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>Date: 12 May, 1998 6:02 PMSubject: Re: glibc2/jdk1.1.5v7I belive glibc2.0.7-7 is in the ftp://ftp.redhat.com/pub/contirb/hurricane/i386I would check but I am useing MacOS (not mine) and it is so messed up that I can no connect to some servers on the internet.>>OK... obviously I've not got them working together, so before I launch>into full diagnostics: I notice on the website that glibc-2.0.7-7 is the>redhat package recommended. Now, not being an rpm type, I only have 2.0.6>(this is the latest one I can *find*) which I compiled myself... So my>question is, _could_ this be the reason for the segfaulting?>>Brief description: sounds a bit like what our friend Travis Shirk was>going through; i.e., "java Class" segfaults, as does "javac". I've done a>fair amount of tweaking and diagnostic work which I can post if anyone>cares...>>I'd appreciate any hints/clues/suggestions>>andy>>->Andy Stubbs +44 151 794 4045>[EMAIL PROTECTED]>>__Get Your Private, Free Email at http://www.hotmail.com
Unidentified subject!
PLEASE DELETE THE INCOGNITO.NET
DOMAIN FROM THIS LIST.
PLEAE ADVISE HOW IT BECAME
LISTED.
-Original Message-From:
Louis-David Mitterrand <[EMAIL PROTECTED]>To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>; [EMAIL PROTECTED]
<[EMAIL PROTECTED]>Date:
12 May, 1998 6:12 PMSubject: Re: Dynamic loading and libgcc
need?On Tue, May 12, 1998 at 10:35:14AM -0700, Robert
Coie wrote:> Is it possible to specify a path for
System.loadLibrary()? I have not> been able to get
LD_LIBRARY_PATH to have any effect. I hacked a> version of
java.lang.Runtime that dumps the result of the>
initializeLinkerInternal call to stderr and it appears to be>
consistently empty regardless of the value of
LD_LIBRARY_PATH.System.load("/path/to/library.so");Regarding
your LD_LIBRARY_PATH problems, are you sure your environmenthas it? If
(for instance) you run your program under some shell spawnedby your
editor you need to make sure the shell reads the proper initfile for
_non-interactive_ use. For example zsh only reads .zshenv whenused from
an editor (that's where I keep CLASSPATH,
LD_LIBRARY_PATH,etc..).> I have also been having problems
using javah when .zip files are in> the classpath, entailing
complaints about inability to resolve various> symbols such as
__muldi3. I suspect this is a result of not having > -lgcc in
the build line when building the shared libraries (notably> libjava
and libzip). I have not been able to test this theory, not>
having access to the original JDK source. ldd -r on libjava.so>
complains about missing libgcc symbols on both my Debian 1.3.1
(libc5> 5.4.33) and pre-2.0 (libc6 2.0.6).Never had this
problem, despite using javah regularly (Debian-2.0). Somelibrary version
mismatch probably.-- Louis-David Mitterrandhttp://www.aparima.com[EMAIL PROTECTED]
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998, Andy Stubbs wrote: > As it happens, it isn't actually there; it's in the updates directory for > RH 5.0. BUT... I can't work out why jdk1.1.5 is linked against a version > of glibc2 which doesn't appear to exist in anything other than .rpm format > and isn't even at the gnu ftp site. This is something of a mystery to me. > I guess if I want this thing to work I'm going to either have to find a > rpm depackager thingy and try to botch the result into the system > somewhere without breaking anything, or give up and get libc5.4.44... > which defeats the whole point somewhat. > > is a .tgz available _anywhere_ (archie thinks not)? Of this particular package? No. JDK uses this version of glibc because it was what I had on my machine when I compiled it. IMO, this is a much better alternative than being built against the versions of 2.0.7 that I compiled myself. I upgraded to RH's glibc because it contains some bugfixes over the stock 2.0.7 pre-releases and I can say grab the standard RH glibc rather than grab my custom glibc build that hasn't been tested anywhere but on my box. Last I checked, 2.0.7 is still a pre-release as far as GNU is concerned. This is why you won't find it at ftp.gnu.org. RH obviously had their reasons for choosing to use it now. Security fixes was one of those. If you want to build it yourself, grab the .src.rpm from ftp://ftp.redhat.com/pub/redhat/redhat-5.0/updates/SRPMS/. Yes, it's still in rpm format, but it contains the pre 2.0.7 tars and the RH patches. The alien program should help you convert it from rpm to whatever format you like. - cls
apache 1.2.4 + mod-jserv + SdK
Hi, I was wondering if anyone out there is trying the same as I am and that is to set up the apache webserver for linux with the servlet module and development kit voor java servlets. I've been trying to do some things with the new jserv release of java and get it recompiled into a new make of apache 1.2.4 for linux, but unsuccesful sofar. If anyone knows how to get it to work I woul;d greatly appreciate it. I'm running RedHat 5 now, but I can reinstall into Slackware anytime , if that will work. About the Jdk1.1.5v7 thing, about the only thing I've done with it sofar is run ICQJava with it and it works great, without any problems, with RedHat 5. Hans
jdk1.1.5: Using jdb
Hello, I've been trying to get jdb running under jdk1.1.5, with no luck. Has anybody gotten this working? When I run jdb Layout, from the JDE/Debug option, it says: "Initializing jdb..." and nothing more. Has anybody gotten this to work? Thanks, Larry
Re: glibc2/jdk1.1.5v7
On Tue, 12 May 1998 [EMAIL PROTECTED] wrote: > > Just in case anyone hasn't said it yet, add my name to the list of > the eternally grateful for the glibc jdk, Chris,Steve, Karl et al. > I don't normally post "me too"s, but this is gonna be the exception to the rule. Me Too. I've finally got it working. I grabbed the "rpm" thingies from redhat and the source rpm and compiled libc-2.0.7-7 to my own taste... Sounds simple doesn't it? Deceptively so. Big Thanks to everyone who emailed me - I never suspected I'd generate quite so much traffic on this one, Andy - Andy Stubbs +44 151 794 4045 [EMAIL PROTECTED]
