[daip] imagr problem

Eric Greisen egreisen at nrao.edu
Fri Jan 29 12:04:03 EST 2010


Oozeer Nadeem wrote:
> Hi
> 
> I have just upgraded from 31DEC08 to 31DEC09 using the binary 
> installation ... However, I am getting some error message which am not 
> able to figure out .. Can you please help me out
> 
> The imagr error follows:
> 
> hostna> task #: Message
> ----------------------------------------
> LOCALH> IMAGR1: Task IMAGR  (release of 31DEC09) begins
> LOCALH> IMAGR1: Doing no flagging this time
> LOCALH> IMAGR1: ZTXOP2: using translated file name =
> LOCALH> IMAGR1: ZTXOP2: /opt/aips/FITS/1925-296-GMRT.BOX
> LOCALH> IMAGR1: FIELDF: Final field parameters and # read are:
> LOCALH> IMAGR1: Field    1  500  500 -2.7149E+03  1.0385E+03          0
> LOCALH> IMAGR1: Field    2  500  500 -2.6651E+01  5.7446E+00          0
> LOCALH> IMAGR1: Field    3  118  118 -2.7141E+03  1.0376E+03          0
> :
> :
> :
> :
> :
> :
> LOCALH> IMAGR1: Field   24  118  118  1.4024E+04 -7.5036E+03          0
> LOCALH> IMAGR1: Field   25  118  118 -2.0098E+04  8.7612E+03          0
> LOCALH> IMAGR1: ZTXOP2: using translated file name =
> LOCALH> IMAGR1: ZTXOP2: /opt/aips/FITS/1925-296-GMRT.BOX
> LOCALH> IMAGR1: Doing no flagging this time
> LOCALH> IMAGR1: Doing no flagging this time
> LOCALH> IMAGR1: Create 1925-296    .IMAGR .   1 (UV)  on disk  1  cno   52
> LOCALH> IMAGR1: Beginning channel    1 through    1 with  1 IFs
> LOCALH> IMAGR1: IMACPY: Copied    11095 visibilities to be imaged
> LOCALH> IMAGR1: QINIT: did a GET  of      5120 Kwords, OFF   35057949674045
> LOCALH> IMAGR1: UVWAIT: begin finding uniform weights
> LOCALH> IMAGR1: UVWAIT: Average grid weight  2.923E+03
> LOCALH> IMAGR1: UVWAIT: Adding temperance S  5.846E+07
> LOCALH> IMAGR1: UVWAIT: begin applying uniform or other weights
> LOCALH> IMAGR1: UVWAIT: Sum of weights in  1.010E+06 and out  1.010E+06
> LOCALH> IMAGR1: UVWAIT: Noise is increased by a factor 1.000 due to 
> weighting
> LOCALH> IMAGR1: UVWAIT: Average summed weight  2.923E+03 over      11095 vis
> LOCALH> IMAGR1: Create 1925-296    .IBM001.   1 (MA)  on disk  1  cno   53
> LOCALH> IMAGR1: Create 1925-296    .IBM002.   1 (MA)  on disk  1  cno   54
> LOCALH> IMAGR1: Create 1925-296    .IBM003.   1 (MA)  on disk  1  cno   55
> :
> :
> :
> :
> LOCALH> IMAGR1: Create 1925-296    .IBM022.   1 (MA)  on disk  1  cno   74
> LOCALH> IMAGR1: Create 1925-296    .IBM023.   1 (MA)  on disk  1  cno   75
> LOCALH> IMAGR1: ZCREAT ERROR     1 FOR 1925-296    .IBM024.   1
> LOCALH> IMAGR1: CLNOPN: ERROR OPENING CLEAN process object
> LOCALH> IMAGR1: Destroyed  1 extension files of type AN
> LOCALH> IMAGR1: Destroyed  1 extension files of type FQ
> LOCALH> IMAGR1: Destroyed UV image file: catno=     52 disk= 1
> LOCALH> IMAGR1: Purports to die of UNNATURAL causes
> LOCALH> IMAGR1: localhos 31DEC09 TST: Cpu=      0.5  Real=      1  
> IO=        2
> 
> ----------
> 
> I have +rw permissions in the directories and also tried to run the same 
> uv file with same input into the 31DEC08, it seems to work fine

The error code (1) means that the file it is trying to create already 
exists.  That does not mean that there is a file logically named 
1925-296.IBM024.1 but instead a disk file in your first data area named
MAD024001.uuu; assuming that slot 76 (extended hex 24) was the target 
slot.  This can happen is bad things happens to a task or aips such that 
a file gets partly created but is not successfully registered in the 
catalog.  Try mcat and ucat on disk 1 to see if there is something in 
slot 76 or not.  If it appears empty, go to your first data area and 
look for MAD024001*.  If there is such a file or perhaps a CBD024001.* 
file (the header) and nothing in the catalog, simply delete them.  To 
determine what your user number is in extended hex type EHEX nnn inside 
aips where nnn is yor decimal user number.

Why the one version works and the other doesn't is a mystery with the 
only obvious explanation that the 08 did not try to create an image in 
the empty catalog slot that had this lurking and show stopping problem.

ERic Greisen




More information about the Daip mailing list