September 2002 - openSUSE Programming - openSUSE Mailing Lists

can't find -lGLw library
by Stewart Taylor 22 Sep '02

22 Sep '02

Hi All I'm just working through Mark Kilgards OpenGl Programming for the X window system and when I try to compile some of the examples I get the following message. unable to find -lGLw, does anyone know how to resolve this please. TIA Stewart

3 4

Re: [suse-programming-e] What kills a service process during boot ?
by Ofer Biran 17 Sep '02

17 Sep '02

I tried the highest Snn (24) in rc3.d with same problem. (And anyway what an echo & sleep script needs) >From: "Jerry Feldman" <gerald.feldman(a)hp.com> >To: suse-programming-e(a)suse.com >Subject: Re: [suse-programming-e] What kills a service process during boot >? >Date: Tue, 17 Sep 2002 16:46:30 -0400 > >The time it is run from startup is very important. You must make sure that >any dependent communications service have been run successfully prior to >the running of your service. This is done via the Snn symlink. The higher >the number the later on boot that this will start. > >On 17 Sep 2002 at 22:32, Ofer Biran wrote: > > > I have a java service set exactly according to init.d/skelton (this is >on > > Suse 7.2 for 390). When init.d/myservice is run outside boot - it's OK. >When > > run during boot - it starts OK (get to the printout of listen on > > ServerSocket)- but then it is killed ! > > > > Then... I tried a very simple shell script service that just echo > > (i=1,2,3...) and sleep between. It works of course forever outside boot >- > > but in boot, it prints (into a file - I used startproc -l) till 3 and >killed > > !! > > > > So what's the story here ??? I tried all combination of startproc / >direct > > exec / various rc.3/Sxxmyservice - no change - the process starts OK but > > doesn't survive the boot. It's also reported among failed services in >run > > level 3 - but [desparately] I added sleep after the startproc - then it >runs > > some more... and killed... just this time without being reported among >the > > failed services. > > > > Any help / idea will be appreciated ! > > > > Ofer Biran > > > > > > > > > > _________________________________________________________________ > > Send and receive Hotmail on your mobile device: http://mobile.msn.com > > > > > > -- > > To unsubscribe, email: suse-programming-e-unsubscribe(a)suse.com > > For additional commands, email: suse-programming-e-help(a)suse.com > > Archives can be found at: http://lists/archive/suse-programming-e > > > > >-- >Jerry Feldman <gerald.feldman(a)hp.com> >Enterprise Systems Group >Hewlett-Packard Company >200 Forest Street MRO1-3/F1 >Marlboro, Ma. 01752 >508-467-4315 http://www.testdrive.compaq.com/linux/ > > >-- >To unsubscribe, email: suse-programming-e-unsubscribe(a)suse.com >For additional commands, email: suse-programming-e-help(a)suse.com >Archives can be found at: http://lists/archive/suse-programming-e _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com

1 0

RE: [suse-programming-e] What kills a service process during boot ?
by Jonathan Cowherd 17 Sep '02

17 Sep '02

You may want to look at what is needed by an non-interactive shell. Since stdout or stderr isn't redirected, the process is probably returning an error to your JVM which then falls over. You may need to wrap "nohup" around it, but for it to be non-interactive, you'll need to find out what the shell environment needs during boot up. If bash is your shell, there are a lot of nice things to do with bash that are good for those types of things. The basic "sh" shell isn't as flexible. If you are calling that directly, you'll need to figure out what it needs (which is probably redirecting stdout, stdin, and stderr). You'll have the same problem if you try to set something in the background and it attempts to write to stdout, stdin, or stderr. The background process will usually stop (unless you have your shell setup differently). So have your /etc/init.d/mservice script setup the environment then call the java program with all of the stdin, stdout, and stderr accounted for. You may want to look at using Log4J to account for your printing out to STDOUT or a log file, etc. I used to have that problem adding things to my /etc/init.d/boot.local back in the days. Jonathan Paul Cowherd Linux and Java Administrator Genscape, Inc. Email: jonathan.cowherd(a)genscape.com Office: (502) 583-3730 Mobile: (502) 314-0444 -----Original Message----- From: Jerry Feldman [mailto:gerald.feldman@hp.com] Sent: Tuesday, September 17, 2002 4:47 PM To: suse-programming-e(a)suse.com Subject: Re: [suse-programming-e] What kills a service process during boot ? The time it is run from startup is very important. You must make sure that any dependent communications service have been run successfully prior to the running of your service. This is done via the Snn symlink. The higher the number the later on boot that this will start. On 17 Sep 2002 at 22:32, Ofer Biran wrote: > I have a java service set exactly according to init.d/skelton (this is > on > Suse 7.2 for 390). When init.d/myservice is run outside boot - it's OK. When > run during boot - it starts OK (get to the printout of listen on > ServerSocket)- but then it is killed ! > > Then... I tried a very simple shell script service that just echo > (i=1,2,3...) and sleep between. It works of course forever outside boot - > but in boot, it prints (into a file - I used startproc -l) till 3 and killed > !! > > So what's the story here ??? I tried all combination of startproc / > direct > exec / various rc.3/Sxxmyservice - no change - the process starts OK but > doesn't survive the boot. It's also reported among failed services in run > level 3 - but [desparately] I added sleep after the startproc - then it runs > some more... and killed... just this time without being reported among the > failed services. > > Any help / idea will be appreciated ! > > Ofer Biran > > > > > _________________________________________________________________ > Send and receive Hotmail on your mobile device: http://mobile.msn.com > > > -- > To unsubscribe, email: suse-programming-e-unsubscribe(a)suse.com > For additional commands, email: suse-programming-e-help(a)suse.com > Archives can be found at: http://lists/archive/suse-programming-e > -- Jerry Feldman <gerald.feldman(a)hp.com> Enterprise Systems Group Hewlett-Packard Company 200 Forest Street MRO1-3/F1 Marlboro, Ma. 01752 508-467-4315 http://www.testdrive.compaq.com/linux/ -- To unsubscribe, email: suse-programming-e-unsubscribe(a)suse.com For additional commands, email: suse-programming-e-help(a)suse.com Archives can be found at: http://lists/archive/suse-programming-e

1 0

What kills a service process during boot ?
by Ofer Biran 17 Sep '02

17 Sep '02

I have a java service set exactly according to init.d/skelton (this is on Suse 7.2 for 390). When init.d/myservice is run outside boot - it's OK. When run during boot - it starts OK (get to the printout of listen on ServerSocket)- but then it is killed ! Then... I tried a very simple shell script service that just echo (i=1,2,3...) and sleep between. It works of course forever outside boot - but in boot, it prints (into a file - I used startproc -l) till 3 and killed !! So what's the story here ??? I tried all combination of startproc / direct exec / various rc.3/Sxxmyservice - no change - the process starts OK but doesn't survive the boot. It's also reported among failed services in run level 3 - but [desparately] I added sleep after the startproc - then it runs some more... and killed... just this time without being reported among the failed services. Any help / idea will be appreciated ! Ofer Biran _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com

2 1

Fortran 95
by Maura Edelweiss Monville 04 Sep '02

04 Sep '02

I was suggested to look at the web site "www.polyhedron.com" for a Fortran compiler. Actually what is available at that web site is a rigorous comparison of different Fortran compilers. This is good to know but does not solve my problem ... which is... I have to port a Kinetic Monte Carlo code, written in FOrtran 95 (I ignore the platform), to my DELL SusE computer. I know already it will NOT compile with fcc (the GNU compiler) as somone else has already struggled with it and eventually has given up. The ideal situation would be to have a Fortran 95 compiler that runs on SuSE 8.0 . The suggested web site does not tell me if such a compiler does exists for my platform and who sells it (as a Physics graduate student the cheaper the better ... of course). Mauede

4 3

Fortran compiler
by Maura Edelweiss Monville 02 Sep '02

02 Sep '02

I have to install a simulation code on my SuSE 8.0 computer. Such a program was written in Fortran 95.Questions: -------- 1) Is a good Fortran compiler available in the SuSE environment ? I remember some time ago the only available Fortran compiler on Red Hat was actually a pre-processor that translated Fortran inot C code and then the gcc compiler was called to produce executable code ... awful ! 2) Are the SuSE debuggers (especially GVD) available for Fortran code as well ? Thank you in advance. Mauede

1 0

Trouble make'ing gcc
by David List 02 Sep '02

02 Sep '02

I have recently upgraded my gcc 2.95.3. Well, I have kept the 2.95.3 and installed 3.0.4 under /usr/local/, since removing or overwriting the SuSE Linux 7.3 default gcc 2.95.3 would probably break some of the 7.3 packages. Of course I would have liked to have the new 3.2, but the make processes of both 3.1.1 and 3.2 failed on me. There seems to be a problem with libgnat. I enclose the last part of a make of 3.1.1: //////////////////////////////////////////////////////////////////// gnatgcc -c -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -W -Wall -gnatpg -gnata -I- -I. -I../../gcc-3.1.1/gcc/ada ../../gcc-3.1.1/gcc/ada/back_end.adb gnatgcc -c -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -W -Wall -gnatpg -gnata -I- -I. -I../../gcc-3.1.1/gcc/ada ../../gcc-3.1.1/gcc/ada/gnat1drv.adb gnatbind -C -I- -I. -I../../gcc-3.1.1/gcc/ada -o b_gnat1.c -n gnat1drv.ali gnatbind: error while loading shared libraries: libgnat-3.13p.so.1: cannot open shared object file: No such file or directorymake[2]: *** [b_gnat1.c] Error 127 make[2]: Leaving directory `/home/david/FFN/build/gcc/ada' make[1]: *** [gnat1] Error 2 make[1]: Leaving directory `/home/david/FFN/build/gcc' make: *** [all-gcc] Error 2 //////////////////////////////////////////////////////////////////// And here is what I get when searching for the messing file: //////////////////////////////////////////////////////////////////// dt:~ # find / -name libgnat* /usr/lib/gcc-lib/i386-Ada-linux/2.8.1/adalib/libgnat.a /usr/lib/gcc-lib/i386-Ada-linux/2.8.1/adalib/libgnat.so /usr/lib/gcc-lib/i386-Ada-linux/2.8.1/adalib/libgnat_so.a /usr/lib/gcc-lib/i386-Ada-linux/2.8.1/adalib/libgnat-3.13p.so /usr/lib/gcc-lib/i386-Ada-linux/2.8.1/libgnatgcc.a /usr/lib/libgnat-3.15a.so /usr/i486-linux-libc5/lib/libgnat.so.3.09 /usr/i486-linux-libc5/lib/libgnat.so //////////////////////////////////////////////////////////////////// I have the impression that libgnat is part of the whole gcc shebang and should be built with all the other stuff when you build the whole of gcc. Am I wrong in this? Should I upgrade libgnat by itself first? Best regards, David List

2 2