Welcome to iraf.net Friday, May 10 2024 @ 04:45 AM GMT
german |
01/09/2010 09:21PM (Read 3619 times)
|
|
|
Status: offline
Registered: 10/02/2008
Posts: 11
|
Hello mike:I've installed IRAF in a AMD 64-bit Athlon machine, but I've the following problem. When I enter cl without doing mkiraf, I get the following:[german@centauri ~]$ cl
Warning: no login.cl found in login directory
dataio images lists obsolete proto system
dbms language noao plot softools utilities
ecl> But now I do mkiraf in some directory (terminal type: xterm) and enter cl:[german@centauri iraf]$ cl
setting terminal type to xterm...
ERROR: Abnormal termination of child process '??'
called as: `prcache (directory)'
called as: `cl ()'
called as: `cl ()'
Error while reading login.cl file - may need to rebuild with mkiraf
Fatal startup error. CL dies.Rebuilding with mkiraf several times doesn't solve the issue.Can you help me?Cheers,
Germán.P.S. I'm using ecl.e from https://iraf.net/ftp/pub/fitz/ecl.e
|
|
|
|
fitz |
01/09/2010 09:21PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
Because the CL binary seems to start okay, my guess is that the x_system.e binary is the cause of the 'abnormal termination' message. This usually means there is e.g. a missing (32-bit compatibility?) shared library or a permissions problem. You should go to the $iraf/bin.redhat directory at the unix prompt and try the following[code:1:49ecfaf9cb]
% ldd x_system.e
% ./x_system.e
[/code:1:49ecfaf9cb]The first command will list the required shared libs, look for 'not found' messages. The second will run the binary directly, if it works you'll see a '>' prompt (type 'bye' to exit) otherwise the host error message might help.-Mike
|
|
|
|
german |
01/09/2010 09:21PM
|
|
|
Status: offline
Registered: 10/02/2008
Posts: 11
|
Hi Mike:Thanks for your prompt reply. This is what you asked for:[code:1:29baa4ae82][german@centauri bin.redhat]$ ldd x_system.e
not a dynamic executable[/code:1:29baa4ae82][code:1:29baa4ae82][german@centauri bin.redhat]$ ./x_system.e
bash: ./x_system.e: /lib/ld-linux.so.2: bad ELF interpreter: No such file or directory[/code:1:29baa4ae82]I forgot to mention that I'm running Fedora 12 x86_64.
Germán.
|
|
|
|
fitz |
01/09/2010 09:21PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
You could try[code:1:a71b61213a]% yum install elfutils elfutils-libelf elfutils-libs elfutils-libelf-devel[/code:1:a71b61213a]but basically the binary is expecting something that will run the older ELF binaries, I can't say whether this is limited to the interpreter or some 32-bin compatibility issue. Also check whether you've got all possible glibc i386 packages installed, Fedora 12 is really new compared to the system used to build the binaries.
|
|
|
|
german |
01/09/2010 09:21PM
|
|
|
Status: offline
Registered: 10/02/2008
Posts: 11
|
Hi Mike:Installing [b:4ab649bbdd]glibc.i686[/b:4ab649bbdd] did the trick![code:1:4ab649bbdd][german@centauri bin.redhat]$ ldd x_system.e
linux-gate.so.1 => (0xf773c000)
libm.so.6 => /lib/libm.so.6 (0xf76f3000)
libc.so.6 => /lib/libc.so.6 (0xf757e000)
/lib/ld-linux.so.2 (0xf773d000)[/code:1:4ab649bbdd][code:1:4ab649bbdd][german@centauri bin.redhat]$ ./x_system.e
>
> bye[/code:1:4ab649bbdd]Now I can run CL perfectly in Fedora 12 x86_64.Thank you very much for your help.Germán.
|
|
|
|
| |
|
Content generated in: 0.16 seconds |
|