iraf.net

Hi, I'm working for my Laurea in padova, and I'm an Iraf and linux beginner,now I have a problem:
I tried to use a simple script which execute hedit and scombine only, and the first time it worked well, but during the process a blackout turned off my pc.
So I tried to rerun the script but now iraf give me this error:
*** glibc detected *** /iraf/iraf/bin.redhat/x_images.e: malloc(): memory corruption: 0x0a4b6a68 ***
======= Backtrace: =========
/lib/libc.so.6[0xbbf0f2]
/lib/libc.so.6(malloc+0x73)[0xbc07f4]
/iraf/iraf/bin.redhat/x_images.e[0x82a6e71]
/iraf/iraf/bin.redhat/x_images.e[0x827afae]
/iraf/iraf/bin.redhat/x_images.e[0x828d25f]
/iraf/iraf/bin.redhat/x_images.e[0x828629a]
/iraf/iraf/bin.redhat/x_images.e[0x8285ff6]
/iraf/iraf/bin.redhat/x_images.e[0x8255a60]
/iraf/iraf/bin.redhat/x_images.e[0x82364d6]
/iraf/iraf/bin.redhat/x_images.e[0x8219ef5]
/iraf/iraf/bin.redhat/x_images.e[0x82178eb]
/iraf/iraf/bin.redhat/x_images.e[0x81fd85f]
/iraf/iraf/bin.redhat/x_images.e[0x829fbfd]
/iraf/iraf/bin.redhat/x_images.e[0x82179a7]
/iraf/iraf/bin.redhat/x_images.e[0x81db927]
/iraf/iraf/bin.redhat/x_images.e[0x80a4f5a]
/iraf/iraf/bin.redhat/x_images.e[0x809d0e6]
/iraf/iraf/bin.redhat/x_images.e[0x804aa68]
/iraf/iraf/bin.redhat/x_images.e[0x8275f60]
/iraf/iraf/bin.redhat/x_images.e[0x81ca60c]
/lib/libc.so.6(__libc_start_main+0xdc)[0xb704e4]
/iraf/iraf/bin.redhat/x_images.e(sinh+0x5d)[0x804a201]
======= Memory map: ========
001a3000-001ae000 r-xp 00000000 fd:00 720947 /lib/libgcc_s-4.1.1-20060525.so.1
001ae000-001af000 rwxp 0000a000 fd:00 720947 /lib/libgcc_s-4.1.1-20060525.so.1
00b3e000-00b57000 r-xp 00000000 fd:00 720898 /lib/ld-2.4.so
00b57000-00b58000 r-xp 00018000 fd:00 720898 /lib/ld-2.4.so
00b58000-00b59000 rwxp 00019000 fd:00 720898 /lib/ld-2.4.so
00b5b000-00c88000 r-xp 00000000 fd:00 720914 /lib/libc-2.4.so
00c88000-00c8a000 r-xp 0012d000 fd:00 720914 /lib/libc-2.4.so
00c8a000-00c8b000 rwxp 0012f000 fd:00 720914 /lib/libc-2.4.so
00c8b000-00c8e000 rwxp 00c8b000 00:00 0
00c90000-00cb3000 r-xp 00000000 fd:00 720931 /lib/libm-2.4.so
00cb3000-00cb4000 r-xp 00022000 fd:00 720931 /lib/libm-2.4.so
00cb4000-00cb5000 rwxp 00023000 fd:00 720931 /lib/libm-2.4.so
08048000-082b2000 r-xp 00000000 fd:00 22939685 /iraf/irafbin/bin.redhat/x_images.e
082b2000-08302000 rwxp 00269000 fd:00 22939685 /iraf/irafbin/bin.redhat/x_images.e
08302000-0843a000 rwxp 08302000 00:00 0
0a433000-0a51e000 rwxp 0a433000 00:00 0
40000000-40001000 r-xp 40000000 00:00 0 [vdso]
40001000-40002000 rwxp 40001000 00:00 0
40019000-4001a000 rwxp 40019000 00:00 0
4005a000-40063000 r-xp 00000000 fd:00 720936 /lib/libnss_files-2.4.so
40063000-40064000 r-xp 00008000 fd:00 720936 /lib/libnss_files-2.4.so
40064000-40065000 rwxp 00009000 fd:00 720936 /lib/libnss_files-2.4.so
40100000-40121000 rwxp 40100000 00:00 0
40121000-40200000 --xp 40121000 00:00 0
bf85f000-bf874000 rwxp bf85f000 00:00 0 [stack]and iraf block itself in this mode.
I use fedora core 5, and I have no .cshrc but .tcshrc (created by me) in my home folder.what should I do?
Thanks
Andrea

Andrea,Two thoughts come to mind: Since you mention a power blackout the possibility that a memory chip was damaged is not out of line. You can use the unix /bin/dmesg command to check the system logs for reports about errors and Google will turn up a number of programs than can be used to diagnose hardware problems. Likewise the image itself may have been corrupted: Does HEDIT fail (since the error comes from the x_images.e binary) fail for all images or just this one? Still, something like this is rare, but if "it worked before the blackout" then it shouldn't be overlooked.For recent 2.6 linux kernels a more common problem is a reported "Memory has been corrupted" error from all tasks (e.g. 'help implot' would do it). In recent release this should be fixed automatically by having the 'cl' command issue a "limit stacksize unlimited" command when it starts up. If everything in iraf seems to fail, type this at the unix prompt before logging into the CL to see if that fixes it.Hope this helps,
-Mike

But everything else run! and in particular Hedit, infact that is the first part of the script. the script is the following:
procedure rebin_spec(input,inini,outre,op,wini,wfin,dw)file input = "" {prompt = "List of images to combine"}
file inini = "" {prompt = "Name of the first image to combine"}
file outre = "" {prompt = "Name of the combined image"}
int op = "" {prompt = "Number of lines in the spectra"}
real wini = "" {prompt = "Starting wavelength of output spectra (#/INDEF)"}
real wfin = "" {prompt = "Ending wavelength of output spectra (#/INDEF)"}
real dw = "" {prompt = "Wavelength increment of output spectra (#/INDEF)"}
beginint ndim
imcopy (inini,outre, verbose=yes)hedit (image=outre, field="CRVAL1", value=wini, add=yes, addonly=no,
delete=no, verify=no, update=yes)
hedit (image=outre, field="CDELT1", value=dw, add=yes, addonly=no,
delete=no, verify=no, update=yes)
hedit (image=outre, field="CD1_1", value=dw, add=yes, addonly=no,
delete=no, verify=no, update=yes)noao
onedspecfor (ndim = 1; ndim <=op; ndim+=1) { scombine ("@"//input,
"borrar.fits", noutput="", logfile="STDOUT", apertures=ndim, group="apertures",
combine="median", reject="none", first=no, w1=wini, w2=wfin, dw=dw,
nw=INDEF, scale="mode", zero="none", weight="none", sample="",
lthreshold=INDEF, hthreshold=INDEF, nlow=1, nhigh=0, nkeep=1, mclip=yes,
lsigma=3., hsigma=3., rdnoise="0.", gain="1.", snoise="0.", sigscale=0.1,
pclip=-0.5, grow=0, blank=0.) imcopy ("borrar.fits",outre // "[*,"// ndim//"]", verbose=yes)

if (access("borrar.fits")) imdel("borrar.fits",verify=no) }
endI tryed to type "limit stacksize unlimited" in the terminal before typing CL (is this correct?) but nothing happened.
The errore come again, but not always, sometimes only.
And I also checked the hardware, and there is no problem.
What to do?

I'm not sure what else to suggest. Is this is a script that ran reliably before the blackout or not at all since then? Can you say that it only happens with a particular image or type of image (e.g. only images from a particular instrument?)? Can you run the HEDIT command on that image outside of the script and see the same error? Does it fail only when you login through your normal directory or as yourself, or will it work if you create a new login.cl in /tmp and/or login as another user?Certainly HEDIT is used thousands of times a day and we're not aware of any problems. I'll need some way to reproduce the error to track down a bug, the above suggestions might rule out something in your environment or about the image to provide a hint, but your comment about the blackout still sounds suspicious.Cheers,
-Mike

ok, I made a lot of attempts, the last one was:
using hedit alone... and it works;
using scombine alone... and it works;
using imcopy alone (to copy the result of scombine, in a row of the final image)... and it DOESN'T work!!
The error is always the same and I don't know how solve it.
Another thing is important: my images are from FORS2 at VLT, I tried to use my script with the images of chip1, and it works perfectly, the problem is only with the images of chip2 (in my original DVD the images were separated before!)!!Cheers,
Andrea

Andrea,Please send me the example image along with exactly what command you tried that failed, I should be able to track it down quickly if I can reproduce the problem.Cheers,
-Mike

Dear Mike,
thank you for time you spent to help me....
but I resolved the problem: the scombine output was (in the case of chip2) 3 pixel in x longer than the output of imcopy (the first one) so, when the second imcopy tried to copy the output of scombine in that one of imcopy, the different dimension crashed everything.Cheers
Andrea