mscred.mscdisplay and .mscexamine
krust wrote on Dec 16, 2010
I've installed v2.15, 64 bit Linux version. mscdisplay and mscexamine give "segmentation error" immediately after it has prompted for the image name and frame. I'm using DS9, this all worked fine before the upgrade. I'm trying to interact with Mosaic II images. Other tasks in the mscred package seem to work as normal. The parameter files look OK I think. Suggestions?
image = "" root name for image to be displayed
frame = 1 frame to be written into
(zero = "") root name for zero subtraction
(mimpars = "") mosaic image parameters (extensions, gaps, proc
(check = no) check if image is loaded
(onepass = yes) load all extensions in one pass?
(bpmask = "BPM") bad pixel mask
(bpdisplay = "none") bad pixel display (none|overlay|interpolate)
(bpcolors = "red") bad pixel colors
(overlay = "") overlay mask
(ocolors = "green") overlay colors
(erase = yes) erase frame
(border_erase = no) erase unfilled area of window
(select_frame = yes) display frame being loaded
(fill = no) scale image to fit display window
(zscale = yes) display range of greylevels near median
(contrast = 0.25) contrast adjustment for zscale algorithm
(zrange = yes) display full image intensity range
(zmask = "") sample mask
(zcombine = "auto") Algorithm for combining z1 and z2 values for ea
(nsample = 1000) maximum number of sample pixels to use
(order = 0) spatial interpolator order (0=replicate, 1=line
(z1 = 0.) minimum greylevel to be displayed
(z2 = 1000.) maximum greylevel to be displayed
(ztrans = "linear") greylevel transformation (linear|log|none|user)
(lutfile = "") file containing user defined look up table
(mode = "ql")
image = "" root name for image to be displayed
frame = 1 frame to be written into
(zero = "") root name for zero subtraction
(mimpars = "") mosaic image parameters (extensions, gaps, proc
(check = no) check if image is loaded
(onepass = yes) load all extensions in one pass?
(bpmask = "BPM") bad pixel mask
(bpdisplay = "none") bad pixel display (none|overlay|interpolate)
(bpcolors = "red") bad pixel colors
(overlay = "") overlay mask
(ocolors = "green") overlay colors
(erase = yes) erase frame
(border_erase = no) erase unfilled area of window
(select_frame = yes) display frame being loaded
(fill = no) scale image to fit display window
(zscale = yes) display range of greylevels near median
(contrast = 0.25) contrast adjustment for zscale algorithm
(zrange = yes) display full image intensity range
(zmask = "") sample mask
(zcombine = "auto") Algorithm for combining z1 and z2 values for ea
(nsample = 1000) maximum number of sample pixels to use
(order = 0) spatial interpolator order (0=replicate, 1=line
(z1 = 0.) minimum greylevel to be displayed
(z2 = 1000.) maximum greylevel to be displayed
(ztrans = "linear") greylevel transformation (linear|log|none|user)
(lutfile = "") file containing user defined look up table
(mode = "ql")
Mike Fitzpatrick wrote on Dec 16, 2010
When you did the IRAF upgrade, did you also upgrade the MSCRED package or is this the same binary you were using before? If you upgraded MSCRED, did you install the 64-bit binary? If so, was this using the dynamic package mechanism (i.e. a 'make mscred' in iraf$extern) or did you install the package manually? Do you have the MSCDB files installed?
krust wrote on Dec 16, 2010
Yes, yes, and yes dynamic. MSCDB files are there.
It appears to do the "erase" and then crash.with ERROR: segmentation violation.
One time after I set check=yes it calculated the display levels and then crashed with the error message:
PANIC in /iraf/iraf/extern/mscred/bin.linux64/x-mscdiplay.e : Memory fwa has been corrupted
But I have not been able to repeat this failure path.
It appears to do the "erase" and then crash.with ERROR: segmentation violation.
One time after I set check=yes it calculated the display levels and then crashed with the error message:
PANIC in /iraf/iraf/extern/mscred/bin.linux64/x-mscdiplay.e : Memory fwa has been corrupted
But I have not been able to repeat this failure path.
Mike Fitzpatrick wrote on Dec 16, 2010
Could you please post a sample problem image to the anonftp at ftp://iraf.noao.edu/pub ? I'm not able to reproduce the problem here and suspect there may be some header keyword issue causing the problem. It would also help if you could post the output of the command
In particular I'm looking at the file size and whether the files data from Dec 16 or much earlier.
cl> dir mscbin$ long+In particular I'm looking at the file size and whether the files data from Dec 16 or much earlier.
krust wrote on Dec 16, 2010
Well, these are Mosaic II images that displayed OK with mscdisp and ds9 in the past, pre the 64-bit upgrade. I'm presently uploading an image "obj033.fits" into the iraf ftp pub directory as requested.
BTW, " disp image[7] 1" works fine.
Here is the response to "dir msc$bin long+ "
-b-rwr-r- iraf 1409550 Nov 29 15:45 msc$bin$libccdred.a
-b-rwr-r- iraf 1969610 Nov 29 15:45 msc$bin$libcombine.a
-b-rwr-r- iraf 320280 Nov 29 15:45 msc$bin$libfinder.a
-b-rwr-r- iraf 893492 Nov 29 15:45 msc$bin$libimexam.a
-b-rwr-r- iraf 1280214 Nov 29 15:45 msc$bin$libmscdisp.a
-b-rwr-r- iraf 2420296 Nov 29 15:45 msc$bin$libmscred.a
-b-rwr-r- iraf 620716 Nov 29 15:45 msc$bin$libsf.a
xb-rwr-r- iraf 4642887 Nov 29 15:46 msc$bin$x_ccdred.e
-b-rwr-r- iraf 14744 Nov 29 15:46 msc$bin$x_ccdred.o
xb-rwr-r- iraf 4298434 Nov 29 15:46 msc$bin$x_combine.e
-b-rwr-r- iraf 13840 Nov 29 15:45 msc$bin$x_combine.o
xb-rwr-r- iraf 6216801 Nov 29 15:46 msc$bin$x_finder.e
-b-rwr-r- iraf 12320 Nov 29 15:46 msc$bin$x_finder.o
xb-rwr-r- iraf 5335006 Nov 29 15:46 msc$bin$x_mscdisplay.e
-b-rwr-r- iraf 13328 Nov 29 15:46 msc$bin$x_mscdisplay.o
xb-rwr-r- iraf 5441930 Nov 29 15:46 msc$bin$x_mscexam.e
-b-rwr-r- iraf 12392 Nov 29 15:46 msc$bin$x_mscexam.o
xb-rwr-r- iraf 5977056 Nov 29 15:46 msc$bin$x_mscred.e
-b-rwr-r- iraf 20504 Nov 29 15:46 msc$bin$x_mscred.o
BTW, " disp image[7] 1" works fine.
Here is the response to "dir msc$bin long+ "
-b-rwr-r- iraf 1409550 Nov 29 15:45 msc$bin$libccdred.a
-b-rwr-r- iraf 1969610 Nov 29 15:45 msc$bin$libcombine.a
-b-rwr-r- iraf 320280 Nov 29 15:45 msc$bin$libfinder.a
-b-rwr-r- iraf 893492 Nov 29 15:45 msc$bin$libimexam.a
-b-rwr-r- iraf 1280214 Nov 29 15:45 msc$bin$libmscdisp.a
-b-rwr-r- iraf 2420296 Nov 29 15:45 msc$bin$libmscred.a
-b-rwr-r- iraf 620716 Nov 29 15:45 msc$bin$libsf.a
xb-rwr-r- iraf 4642887 Nov 29 15:46 msc$bin$x_ccdred.e
-b-rwr-r- iraf 14744 Nov 29 15:46 msc$bin$x_ccdred.o
xb-rwr-r- iraf 4298434 Nov 29 15:46 msc$bin$x_combine.e
-b-rwr-r- iraf 13840 Nov 29 15:45 msc$bin$x_combine.o
xb-rwr-r- iraf 6216801 Nov 29 15:46 msc$bin$x_finder.e
-b-rwr-r- iraf 12320 Nov 29 15:46 msc$bin$x_finder.o
xb-rwr-r- iraf 5335006 Nov 29 15:46 msc$bin$x_mscdisplay.e
-b-rwr-r- iraf 13328 Nov 29 15:46 msc$bin$x_mscdisplay.o
xb-rwr-r- iraf 5441930 Nov 29 15:46 msc$bin$x_mscexam.e
-b-rwr-r- iraf 12392 Nov 29 15:46 msc$bin$x_mscexam.o
xb-rwr-r- iraf 5977056 Nov 29 15:46 msc$bin$x_mscred.e
-b-rwr-r- iraf 20504 Nov 29 15:46 msc$bin$x_mscred.o
krust wrote on Dec 16, 2010
Sorry Mosaic Image file transfer times out. Several tries. Well, you have lots of Mosaic Imager files in Tucson, it's not as though these ones are any different.
krust wrote on Dec 16, 2010
mscred> mscdisp
root name for image to be displayed (obj048.fits):
frame to be written into (1:16) (1):
Image: Individual Display (zcombine=minmax)
1: 7.6 41.0 -3.7 60.1
3: 2.1 43.6 -3.7 60.1
5: -3.7 54.9 -3.7 60.1
7: 3.1 48.3 -3.7 60.1
9: 6.1 39.0 -3.7 60.1
11: 4.4 54.2 -3.7 60.1
13: 1.7 60.1 -3.7 60.1
15: 1.8 59.5 -3.7 60.1
PANIC in `/iraf/iraf/extern/mscred/bin.linux64/x_mscdisplay.e': Memory fwa has been corrupted
root name for image to be displayed (obj048.fits):
frame to be written into (1:16) (1):
Image: Individual Display (zcombine=minmax)
1: 7.6 41.0 -3.7 60.1
3: 2.1 43.6 -3.7 60.1
5: -3.7 54.9 -3.7 60.1
7: 3.1 48.3 -3.7 60.1
9: 6.1 39.0 -3.7 60.1
11: 4.4 54.2 -3.7 60.1
13: 1.7 60.1 -3.7 60.1
15: 1.8 59.5 -3.7 60.1
PANIC in `/iraf/iraf/extern/mscred/bin.linux64/x_mscdisplay.e': Memory fwa has been corrupted
Mike Fitzpatrick wrote on Dec 16, 2010
Could you try updating the MSCRED package to the latest version, i.e.
And let me know if the problem continues? I think the version you have is and older one, Frank just updated it to v5.0 with his 64-bit changes Dec 16 and I want to make sure we're debugging the same thing.
% cd $iraf/extern
% make mscredAnd let me know if the problem continues? I think the version you have is and older one, Frank just updated it to v5.0 with his 64-bit changes Dec 16 and I want to make sure we're debugging the same thing.
krust wrote on Dec 16, 2010
Here's what happens -
make mscred
Adding dependency 'fitsutil' ....
Installing package 'fitsutil' .... [Error]
Hmm, looks suspicious? In any case, mscdisp still crashes.
make mscred
Adding dependency 'fitsutil' ....
Installing package 'fitsutil' .... [Error]
Hmm, looks suspicious? In any case, mscdisp still crashes.
Mike Fitzpatrick wrote on Dec 16, 2010
You get the 'Error' return when either the package cannot be retrieved from the repository, or it cannot be unpacked in the iraf$extern directory. Either case can be caused by not having write permission on the iraf$extern directory or the files therein. Because the FITSUTIL install failed, MSCRED was never updated so it still fails because you're still using the old binaries.
Do you have the proper write permissions?
Do you have the proper write permissions?
krust wrote on Dec 16, 2010
OK, it needed to be done as IRAF user, so I needed help for that.
iraf@ctiowi> make mscred
Adding dependency 'fitsutil' ....
Installing package 'fitsutil' .... [OK]
Installing package 'mscred' .... [OK]
iraf@ctiowi>
[1] Done xgterm
However the original problem remains:
mscred> mscdisp
root name for image to be displayed (obj048.fits):
frame to be written into (1:16) (1):
Image: Individual Display (zcombine=minmax)
1: 7.6 41.0 -3.7 60.1
3: 2.1 43.6 -3.7 60.1
5: -3.7 54.9 -3.7 60.1
7: 3.1 48.3 -3.7 60.1
9: 6.1 39.0 -3.7 60.1
11: 4.4 54.2 -3.7 60.1
13: 1.7 60.1 -3.7 60.1
15: 1.8 59.5 -3.7 60.1
PANIC in `/iraf/iraf/extern/mscred/bin.linux64/x_mscdisplay.e': Memory fwa has been corrupted
Most often the error message is just "segmentation violation".
iraf@ctiowi> make mscred
Adding dependency 'fitsutil' ....
Installing package 'fitsutil' .... [OK]
Installing package 'mscred' .... [OK]
iraf@ctiowi>
[1] Done xgterm
However the original problem remains:
mscred> mscdisp
root name for image to be displayed (obj048.fits):
frame to be written into (1:16) (1):
Image: Individual Display (zcombine=minmax)
1: 7.6 41.0 -3.7 60.1
3: 2.1 43.6 -3.7 60.1
5: -3.7 54.9 -3.7 60.1
7: 3.1 48.3 -3.7 60.1
9: 6.1 39.0 -3.7 60.1
11: 4.4 54.2 -3.7 60.1
13: 1.7 60.1 -3.7 60.1
15: 1.8 59.5 -3.7 60.1
PANIC in `/iraf/iraf/extern/mscred/bin.linux64/x_mscdisplay.e': Memory fwa has been corrupted
Most often the error message is just "segmentation violation".
Francisco Valdes wrote on Dec 16, 2010
Hi Alistair,
I'm sorry you've been having problems. I am looking at your exchanges. The first thing to report is that I don't have a problem. I got a random Mosaic2 exposure and displayed the raw file. Then I noticed that the output shows just the odd numbers which means you ran ccdproc and merged the amplifier pairs. So I did that and it still displays ok. So what we have to determine is a) that you indeed have the same version I am using (which is the latest) and b) that it is not data dependent. Could you repeat the
command again? Could you also send a long header listing of the problem image (just the first extension will do):
This will tell me what you've done so far and also will allow me to get a copy of the same raw file from the archive.
It might also be that you need to start with the raw file and redo the ccdproc with the latest mscred version.
If you want to take this directly to me through email that would be fine and probably better than posting directory and header listings to the help forum.
Frank Valdes
I'm sorry you've been having problems. I am looking at your exchanges. The first thing to report is that I don't have a problem. I got a random Mosaic2 exposure and displayed the raw file. Then I noticed that the output shows just the odd numbers which means you ran ccdproc and merged the amplifier pairs. So I did that and it still displays ok. So what we have to determine is a) that you indeed have the same version I am using (which is the latest) and b) that it is not data dependent. Could you repeat the
dir mscbin$ l+
command again? Could you also send a long header listing of the problem image (just the first extension will do):
imhead obj048.fits[1] l+
This will tell me what you've done so far and also will allow me to get a copy of the same raw file from the archive.
It might also be that you need to start with the raw file and redo the ccdproc with the latest mscred version.
If you want to take this directly to me through email that would be fine and probably better than posting directory and header listings to the help forum.
Frank Valdes
Francisco Valdes wrote on Dec 16, 2010
I have found a bug in the mscdisplay code that caused a segmentation violation after a display completes. The error is not specifically 64-bit related though apparently it only was obvious with 64-bit use. Please update mscred from the V2.15 repository.
Frank Valdes
Frank Valdes
Last post on Dec 16, 2010