[daip] problem in SNCOR and CLCAL in 31DEC15

[Eric Greisen]

On 11/02/2015 10:47 PM, Hiroshi Imai wrote:
> Dear Eric,
>
> We have used AIPS and appreciate for your great contribution to AIPS
> management and development.
> Today, I would like to ask how to fix our recent problems in SNCOR and
> CLCAL.
> We have analyzed VERA data in AIPS,
> but recently found the problems in one of our machines installing the
> 31DEC15 version.
> We loaded a text file in TBIN and its solutions are displayed properly
> in SNPLT.
> However, it does not work in SNCOR, showing the error messages as follows.
>
> YUTA-U> SNCOR1: Task SNCOR  (release of 31DEC15) begins
> YUTA-U> SNCOR1: TABINI: KEY COUNTS      7     51 OR COLS     17     15
> DON'T MAT
> YUTA-U>         CH
> YUTA-U> SNCOR1: SNINI: ERROR   1 FROM TABINI OPCODE = WRIT
> YUTA-U> SNCOR1: SNINI: ERROR INITIALIZING SOLUTION TABLE FOR WRIT
> YUTA-U> SNCOR1: Purports to die of UNNATURAL causes
> YUTA-U> SNCOR1: yuta-ux31e   31DEC15 TST: Cpu=       0.0  Real=       0
>
> It does not work in CLCAL either, in which no calibration was made to an
> updated CL table.
> Strangely, only one of our machines that install the same AIPS version
> (15DEC15) has such problems.
> Taking into account the error messages described above, our input text
> file may cause  the problems.
> However  I have not idea where is the problem in it through the
> aforementioned error messages.
>

These symptoms confuse me I fear.  The TABINI error above tells me that 
you have code after the format revision to the SN table.  Your text file 
is in the format of revision 10 of the SN table, so SNCOR should
reformat the file on the fly for you.  When I try it with your SN table 
I get
primat> SNCOR1: Task SNCOR  (release of 31DEC15) begins
primat> SNCOR1: SNREFM: Reformatting SN table  32
primat> SNCOR1:  47464 Records modified
primat> SNCOR1: Appears to have ended successfully
primat> SNCOR1: primate      31DEC15 TST: Cpu=       0.0  Real=       0
Run PRTAB on your SN table.  It should show revision 10 after TBIN and 
15 columns.  After SNCOR is should show 17 columns and revision 11.

Perhaps the easiest work-around would be to copy your data set with 
UVCOP.  That should reformat the SN tables to revision 11 with 17 columns.

Eric Greisen