iraf 2.16 hselect does not work well
Robert Janusz wrote on Apr 09, 2012
Here is a code:
works fine. And now
713 MEF FITS images is OK, but more hangs the "hselect" with output:
# kill -9 ....
Really strange...
cd RAW
s1='$I,object,imagetyp,filter,ut,exptime'
hselect("*.fits[0]",s1,yes)works fine. And now
cd ..
ecl> hselect("RAW/*.fits[0]",s1,yes)713 MEF FITS images is OK, but more hangs the "hselect" with output:
*** glibc detected *** /iraf/iraf/bin.linux/x_images.e: malloc(): smallbin double linked list corrupted: 0x0c65e628 ***
======= Backtrace: =========
/lib/i686/cmov/libc.so.6(+0x6b381)[0x400dc381]
/lib/i686/cmov/libc.so.6(+0x6e500)[0x400df500]
/lib/i686/cmov/libc.so.6(__libc_malloc+0x5c)[0x400e0d8c]
/iraf/iraf/bin.linux/x_images.e[0x82e7834]
/iraf/iraf/bin.linux/x_images.e[0x82b96a1]
/iraf/iraf/bin.linux/x_images.e[0x82b97a1]
/iraf/iraf/bin.linux/x_images.e[0x828a799]
/iraf/iraf/bin.linux/x_images.e[0x8288fd7]
/iraf/iraf/bin.linux/x_images.e[0x8203a0d]
/iraf/iraf/bin.linux/x_images.e[0x8202cbc]
/iraf/iraf/bin.linux/x_images.e[0x8202c56]
/iraf/iraf/bin.linux/x_images.e[0x80a78e3]
/iraf/iraf/bin.linux/x_images.e[0x804dcc2]
/iraf/iraf/bin.linux/x_images.e[0x82b3a76]
/iraf/iraf/bin.linux/x_images.e[0x81f3541]
/lib/i686/cmov/libc.so.6(__libc_start_main+0xe6)[0x40087ca6]
/iraf/iraf/bin.linux/x_images.e[0x804d4c1]
======= Memory map: ========
08048000-0849f000 r-xp 00000000 08:07 329977 /iraf/iraf216/bin.linux/x_images.e
0849f000-084f4000 rwxp 00457000 08:07 329977 /iraf/iraf216/bin.linux/x_images.e
084f4000-0ab38000 rwxp 00000000 00:00 0
0b715000-35dfc000 rwxp 00000000 00:00 0 [heap]
40000000-4001b000 r-xp 00000000 08:07 484544 /lib/ld-2.11.3.so
4001b000-4001c000 r-xp 0001b000 08:07 484544 /lib/ld-2.11.3.so
4001c000-4001d000 rwxp 0001c000 08:07 484544 /lib/ld-2.11.3.so
4001d000-4001e000 r-xp 00000000 00:00 0 [vdso]
4001e000-40020000 rwxp 00000000 00:00 0
40031000-40055000 r-xp 00000000 08:07 499135 /lib/i686/cmov/libm-2.11.3.so
40055000-40056000 r-xp 00023000 08:07 499135 /lib/i686/cmov/libm-2.11.3.so
40056000-40057000 rwxp 00024000 08:07 499135 /lib/i686/cmov/libm-2.11.3.so
40057000-4006c000 r-xp 00000000 08:07 499143 /lib/i686/cmov/libpthread-2.11.3.so
4006c000-4006d000 r-xp 00014000 08:07 499143 /lib/i686/cmov/libpthread-2.11.3.so
4006d000-4006e000 rwxp 00015000 08:07 499143 /lib/i686/cmov/libpthread-2.11.3.so
4006e000-40071000 rwxp 00000000 00:00 0
40071000-401b1000 r-xp 00000000 08:07 499153 /lib/i686/cmov/libc-2.11.3.so
401b1000-401b2000 ---p 00140000 08:07 499153 /lib/i686/cmov/libc-2.11.3.so
401b2000-401b4000 r-xp 00140000 08:07 499153 /lib/i686/cmov/libc-2.11.3.so
401b4000-401b5000 rwxp 00142000 08:07 499153 /lib/i686/cmov/libc-2.11.3.so
401b5000-401b9000 rwxp 00000000 00:00 0
401b9000-401d6000 r-xp 00000000 08:07 482387 /lib/libgcc_s.so.1
401d6000-401d7000 rwxp 0001c000 08:07 482387 /lib/libgcc_s.so.1
401fb000-40205000 r-xp 00000000 08:07 499140 /lib/i686/cmov/libnss_files-2.11.3.so
40205000-40206000 r-xp 00009000 08:07 499140 /lib/i686/cmov/libnss_files-2.11.3.so
40206000-40207000 rwxp 0000a000 08:07 499140 /lib/i686/cmov/libnss_files-2.11.3.so
40207000-4020d000 r-xp 00000000 08:07 499155 /lib/i686/cmov/libnss_compat-2.11.3.so
4020d000-4020e000 r-xp 00006000 08:07 499155 /lib/i686/cmov/libnss_compat-2.11.3.so
4020e000-4020f000 rwxp 00007000 08:07 499155 /lib/i686/cmov/libnss_compat-2.11.3.so
4020f000-40222000 r-xp 00000000 08:07 499137 /lib/i686/cmov/libnsl-2.11.3.so
40222000-40223000 r-xp 00012000 08:07 499137 /lib/i686/cmov/libnsl-2.11.3.so
40223000-40224000 rwxp 00013000 08:07 499137 /lib/i686/cmov/libnsl-2.11.3.so
40224000-40226000 rwxp 00000000 00:00 0
40226000-4022e000 r-xp 00000000 08:07 499133 /lib/i686/cmov/libnss_nis-2.11.3.so
4022e000-4022f000 r-xp 00008000 08:07 499133 /lib/i686/cmov/libnss_nis-2.11.3.so
4022f000-40230000 rwxp 00009000 08:07 499133 /lib/i686/cmov/libnss_nis-2.11.3.so
40300000-40321000 rwxp 00000000 00:00 0
40321000-40400000 ---p 00000000 00:00 0
bfd16000-bfd2b000 rwxp 00000000 00:00 0 [stack]
# kill -9 ....
ecl> flpr ; flpr
ERROR: Abnormal termination of child process 'bin$x_images.e'
called as: `flprcache ()'
ecl>Really strange...
Mike Fitzpatrick wrote on Apr 09, 2012
Could you try the command:
If the date on the binary is before April 5 then I'd suggest you try updating using the commands
I've tried your problem on a test case here and it seems to be fixed, there was previously a know issue with the new list templates.
Another thing to test if you continue to have problems is to use the old list template behavior, i.e.
then re-run your test. If it still throws an error I'll try again to reproduce the problem.
cl> dir bin$x_images.e long+If the date on the binary is before April 5 then I'd suggest you try updating using the commands
% cd $iraf
% make latest
I've tried your problem on a test case here and it seems to be fixed, there was previously a know issue with the new list templates.
Another thing to test if you continue to have problems is to use the old list template behavior, i.e.
cl> reset use_new_imt = no ; flpr 0 # zero, not ohthen re-run your test. If it still throws an error I'll try again to reproduce the problem.
Robert Janusz wrote on Apr 09, 2012
The update did not help. The command:
did the job very well.
reset use_new_imt = no ; flpr 0 did the job very well.
Mike Fitzpatrick wrote on Apr 09, 2012
The problem is that with 713 images an internal string buffer was overflowing and not being resized properly. This (and a few other things) has been fixed, you can update to the latest release with
% cd $iraf
% make latest
Robert Janusz wrote on Apr 09, 2012
Thank you. I did the update and will check it. Anyway, during the update I got:
I did not expect any "Error...". Is it OK?
% make latest
Updating IRAF v2.16 to latest release.
Fri Apr 13 15:21:18 CEST 2012
(util/iraf_update -all)
Updating IRAF system ....
iraf_latest: cwd = /iraf/iraf
Updating binaries: linux
Downloading: patch-src.tar.gz Unpacking ...tar (child): /tmp/patch-src.tar.gz: Cannot open: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now
Done.
Downloading: patch.lnux.x86.tar.gz Unpacking ...tar (child): /tmp/patch.lnux.x86.tar.gz: Cannot open: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now
Done.
Update all external packages ...
...
Update Status: SuccessfulI did not expect any "Error...". Is it OK?
Robert Janusz wrote on Apr 09, 2012
I think the update does not download the patches.
Robert Janusz wrote on Apr 09, 2012
I downloaded patches manually, installed them and I was able to re-update without any error with really downloading from internet. But when I entered "cl" I lost the main screen. So I installed (from the http://iraf.noao.edu/) the "Linux 32-bit (191MB)" and extern "mscred,stsdas". The initial screen still is only:
The command:
runs without any error.
Initializing SAMP .... No Hub Available
dataio. images. mscred. plot. stsdas. utilities.
dbms. language. noao. proto. system. vo.
fitsutil. lists. obsolete. softools. tables.
vocl> The command:
hselect('RAW/*.fits[0]',s1,yes,>>'run')runs without any error.
Robert Janusz wrote on Apr 09, 2012
Now, with new installation, I have the 'mscred' problem again:
mscred
# setinstr ... OK
mscred> ccdlist j0062.fits
Warning: Cannot open image (j0062.fits[0])
j0062.fits[im1][1036,2032][ushort][][1][]:T7M671j
j0062.fits[im2][1036,2032][ushort][][2][]:T7M671j
mscred> ccdproc j0062.fits
Warning: Cannot open image (j0062.fits[1])
mscred> display j0062.fits[1] 2 # OK
z1=1715. z2=1762.726
mscred> display j0062.fits[2] 2 # OK
z1=1729. z2=1787.334Mike Fitzpatrick wrote on Apr 09, 2012
In your first report I think the problem was that the patch file didn't download for some reason, however I've verified the files exist in the proper place and are readable, and a 'make latest' on a system here worked fine. Did a '/tmp/patch-src.tar.gz' file already exist that couldn't be overwritten?
The problem with the missing banner page was supposedly fixed earlier but I found the hlib$motd file was still being overwritten. This has been fixed (again) and verified. You can get the hlib$motd file from ftp://iraf.noao.edu/iraf/v216/support/motd.
Lastly, I updated the mscred binaries linked against the most recent system. A "make latest" would update this or else you can do a 'make mscred" from the $iraf/extern directory to manually update it.
If you still have problems downloading, let me know.
The problem with the missing banner page was supposedly fixed earlier but I found the hlib$motd file was still being overwritten. This has been fixed (again) and verified. You can get the hlib$motd file from ftp://iraf.noao.edu/iraf/v216/support/motd.
Lastly, I updated the mscred binaries linked against the most recent system. A "make latest" would update this or else you can do a 'make mscred" from the $iraf/extern directory to manually update it.
If you still have problems downloading, let me know.
Robert Janusz wrote on Apr 09, 2012
Now the update worked fine but there are still problems:
but this is not so urgent to have. The main problem is still with 'mscred' and it blocs any processing of MEF file:
and I am blocked.
vocl> tinfo ds9.xml
ERROR: Cannot open file (cache$.f775517028)
vocl> fcache init
vocl> tinfo ds9.xml
ERROR: Cannot open file (cache$.f775517028)but this is not so urgent to have. The main problem is still with 'mscred' and it blocs any processing of MEF file:
mscred> ccdlist j0062.fits
Warning: Cannot open image (j0062.fits[0])
j0062.fits[im1][1036,2032][ushort][][1][]:T7M671j
j0062.fits[im2][1036,2032][ushort][][2][]:T7M671j
mscred> ccdproc j0062.fits
Warning: Cannot open image (j0062.fits[1])
mscred> display j0062.fits[1] 2 # OK
z1=1715. z2=1762.726
mscred> display j0062.fits[2] 2 # OK
z1=1729. z2=1787.334and I am blocked.
Mike Fitzpatrick wrote on Apr 09, 2012
The error doesn't mean the ds9.xml file can't be read, only that the cache file cannot be opened. Since you did a 'fcache init' already, can you also do a "show cache" and make sure you have write permission in that directory? Does an "fcache list" show anything in the directory?
The ccdproc problem may be a problem coming from the earlier update. Do the binaries date from at least April 11? If you go to the iraf$extern directory and do a "make mscred" to explicitly force the package to be reinstalled, does that change/fix things?
The ccdproc problem may be a problem coming from the earlier update. Do the binaries date from at least April 11? If you go to the iraf$extern directory and do a "make mscred" to explicitly force the package to be reinstalled, does that change/fix things?
Last post on Apr 09, 2012