iraf.net

Hi Jeannette-
We just upgraded our operating system from 4.1 to 4.1.1, and in the
process seem to have screwed up at least some aspects of iraf. I think
this happened because the / and /usr partitions were close to full,
and some things from tere were inadvertantly deleted by our system
managert in the installation. Those have now been restored, and I
am able to get into iraf iooops!
ok. However, it complains when I try to display an image. I get
the error message:
ERROR: Cannot open device (node!imtool,/dev/imt1,800,800)
My guess is that this is pretty simple to fix, but I am at a loss.
Can you help.? I'm going to lunch right now but will be back in
a bit and will check in with you. If it would help to logon to our
system I'll give you a username and password.

Thanks a bunch,
Frank

Hello IRAF Support,
I'm using IRAF to do data reduction at the American Museum of
Natural History in NYC, and I ran into a problem that no one here
seems to be able to help me with. I have a new data set with a bunch
of .fits files which I am trying to do an overscan and trim on, but
everytime I do the ccdproc *.fits command, an ERROR message pops up,
which says "ERROR: Cannot open pixel file (tmp####a)", with the
'####' being a different 4 digit number each time, appearing to get
larger with each try. When I list files (ls) before doing ccdproc,
only the .fits files appear, but when I ls after the ccdproc and
error message, there are random tmp####.imh files in my directory.
I'm pretty confident my parameters are correct, and that I followed
the right procedures for overscan/trim. I also tried ccdproc on
individual .fits files, but it still does not work (although when
done individually, no error message appears). However, when I
display the image again, nothing has changed. I have also tried this
on multiple data sets, so I know the problem is not with one
specific set. I've tried all the obvious solutions I can think of...
any suggestions??Thanks a lot!
Charlene Kuperstein

Hi Charlene,> I have a new data set with a bunch of .fits files which I am trying
> to do an overscan and trim on, but everytime I do the ccdproc *.fits
> command, an ERROR message pops up, which says "ERROR: Cannot open
> pixel file (tmp####a)", with the '####' being a different 4 digit
> number each time, appearing to get larger with each try.My guess is that you have imtype set to the default value which will
generate IRAF format images, not FITS, on output. IRAF images are
stored in two files, a header (.imh) file in the current directory,
and a pixel (.pix) file in the directory pointed to by the "imdir"
environment variable. Ccdproc, of course, has to create a variety
of temporary images as it does its magic. I suspect that your
imdir isn't pointing at a directory that you have write permission
on.The first thing to check is whether you can write a file (any file)
into the imdir directory - do a "show imdir" to see.The next thing you might try if that seems to problem is to force
IRAF to create FITS images. FITS images are stored in a single file
in the current working directory. To force this to happen, type:
"reset imtype = fits" and try ccdproc again. If this works, edit
the reset command into your login.cl or loginuser.cl file.> I also tried ccdproc on individual .fits files, but it still does
> not work (although when done individually, no error message appears).Hmmm. It's possible that ccdproc doesn't use a temporary file in the
case of a single in-place input image.> However, when I display the image again, nothing has changed.I would recommend that you use imstat to compare before-and-after
pixel statistics rather than relying on how the image looks. Depending
on the constrast and such, a processed image may look very similar to
the raw image - perhaps even if you blink them against each other.Rob Seaman
NOAO/IRAF