[daip] CALIB issue?

[Eric Greisen]

SUSAN BILL WIDER wrote:
> Good morning, Eric.
>  
> I'm running 31DEC08 AIPS on a multiple sub-array data set (18 
> sub-arrays).  After calculating solutions for the first 7 sub-arrays, 
> CALIB chokes in sub-array 8 when it writes the SN table ---
>  
>     Writing SN table    1
>     TABIO: BAD LRNO =     253 LIMIT      252
>     TABINDX:TABIO ERROR  2
>  
> Since CALIB seemed to work fine on this dataset as recently as 17JUN08, 
> I alert you to the possibility that something is broken here.
>  
> I'll check the change log later today to see if there are clues there. 
> I'll also experiment with earlier copies of the data.

Actually not I expect.

This is an NX table read issue and it arises because you have allowed 
times in the data set to get very large (10's or 100's of days).  This 
makes comparison of times inaccurate and causes CALIB to look for an 
extra NX table record.  It is generally unwise to let times get so big.

Perhaps I should work on CALIB to survive this but the issue is really 
very much more fundamental (Real*4 for time in days)

Eric Greisen